Spaces:
Running
Running
| window.BUFO_BENCH = {"schema_version": 4, "models": [{"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "meta/muse-image", "name": "muse-image", "run": "2026-09-05-expansion-meta_muse-image-conditioned", "score": 9, "bufo_count": 9, "close_count": 0, "not_bufo_count": 1, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.1, "cost_per_image_usd": 0.01, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.011111111111111112, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": true, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 38.7}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.2}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.9}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.6}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.7}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 43.8}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 31.8}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.3}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 51.8}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-meta_muse-image-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.4}], "counts": {"bufo": 9, "close": 0, "not_bufo": 1, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 44.9}, {"cost_per_image_usd": 0.231827, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.231477, 0.232254], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [2.3147699999999998, 2.32254], "id": "openai/gpt-5.4-image-2", "name": "gpt-5.4-image-2", "run": "2026-09-04-0315-openai_gpt-5.4-image-2-conditioned", "score": 9, "bufo_count": 9, "close_count": 0, "not_bufo_count": 1, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 2.31827, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.25758555555555557, "cost_per_bufo_range_usd": [0.25719666666666663, 0.25806], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 57.9}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.0}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.2}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.1}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.6}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.5}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 31.0}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 52.3}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.7}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0315-openai_gpt-5.4-image-2-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 38.8}], "counts": {"bufo": 9, "close": 0, "not_bufo": 1, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 45.9}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "openai/gpt-image-2", "name": "gpt-image-2", "run": "2026-09-05-expansion-openai_gpt-image-2-conditioned", "score": 9, "bufo_count": 9, "close_count": 0, "not_bufo_count": 1, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.12791, "cost_per_image_usd": 0.012791, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.014212222222222222, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 54.6}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 47.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 40.2}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 53.5}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 43.2}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 51.1}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 37.1}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.7}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 47.3}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-openai_gpt-image-2-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.9}], "counts": {"bufo": 9, "close": 0, "not_bufo": 1, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 47.1}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "recraft/recraft-v4-styles-pro", "name": "recraft-v4-styles-pro", "run": "2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned", "score": 8, "bufo_count": 8, "close_count": 0, "not_bufo_count": 2, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 1.05, "cost_per_image_usd": 0.10500000000000001, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.13125, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.2}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.0}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.4}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 51.4}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.1}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.6}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 38.5}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.7}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.9}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-recraft_recraft-v4-styles-pro-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.9}], "counts": {"bufo": 8, "close": 0, "not_bufo": 2, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 45.8}, {"cost_per_image_usd": 0.07, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.07, 0.07], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [0.7000000000000001, 0.7000000000000001], "id": "x-ai/grok-imagine-image-2.0", "name": "grok-imagine-image-2.0", "run": "2026-09-05-grok-imagine-image-2.0-conditioned", "score": 8, "bufo_count": 8, "close_count": 0, "not_bufo_count": 2, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.7000000000000001, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.08750000000000001, "cost_per_bufo_range_usd": [0.08750000000000001, 0.08750000000000001], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.3}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 30.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 42.9}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 47.3}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 26.5}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 42.0}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 36.6}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.6}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 34.2}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-grok-imagine-image-2.0-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 30.2}], "counts": {"bufo": 8, "close": 0, "not_bufo": 2, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 38.0}, {"cost_per_image_usd": 0.13930066666666666, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.139096, 0.139418], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [1.39096, 1.39418], "id": "google/gemini-3-pro-image", "name": "gemini-3-pro-image", "run": "2026-09-04-0352-google_gemini-3-pro-image-conditioned", "score": 7, "bufo_count": 7, "close_count": 0, "not_bufo_count": 3, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 1.3930066666666665, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.19900095238095236, "cost_per_bufo_range_usd": [0.19870857142857143, 0.19916857142857142], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 49.0}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.0}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.8}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 47.8}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 36.9}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.6}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 35.7}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 53.8}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 37.8}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0352-google_gemini-3-pro-image-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.3}], "counts": {"bufo": 7, "close": 0, "not_bufo": 3, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 44.2}, {"cost_per_image_usd": 0.06789216666666666, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.067887, 0.0678975], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [0.6788700000000001, 0.678975], "id": "google/gemini-3.1-flash-image", "name": "gemini-3.1-flash-image", "run": "2026-09-04-0120-google_gemini-3.1-flash-image-conditioned", "score": 7, "bufo_count": 7, "close_count": 0, "not_bufo_count": 3, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.6789216666666665, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.0969888095238095, "cost_per_bufo_range_usd": [0.09698142857142858, 0.09699642857142857], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 56.1}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 40.2}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.8}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.7}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 51.0}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 42.5}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 32.3}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 49.2}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 37.5}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0120-google_gemini-3.1-flash-image-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.8}], "counts": {"bufo": 7, "close": 0, "not_bufo": 3, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 44.0}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "sourceful/riverflow-v2.5-pro", "name": "riverflow-v2.5-pro", "run": "2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned", "score": 7, "bufo_count": 7, "close_count": 0, "not_bufo_count": 3, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 2.598604, "cost_per_image_usd": 0.2598604, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.37122914285714287, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 51.4}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.6}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 36.7}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 50.1}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 39.1}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.5}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 38.1}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.5}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 46.2}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-sourceful_riverflow-v2.5-pro-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.0}], "counts": {"bufo": 7, "close": 0, "not_bufo": 3, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 43.9}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "bytedance-seed/seedream-5-0-pro", "name": "seedream-5-0-pro", "run": "2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned", "score": 6, "bufo_count": 6, "close_count": 0, "not_bufo_count": 4, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.8999999999999999, "cost_per_image_usd": 0.09, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.15, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 32.3}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.5}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.6}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 45.7}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.5}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.1}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 32.3}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.7}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.2}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-bytedance-seed_seedream-5-0-pro-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.6}], "counts": {"bufo": 6, "close": 0, "not_bufo": 4, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 41.0}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "qwen/qwen-image-3-pro", "name": "qwen-image-3-pro", "run": "2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned", "score": 6, "bufo_count": 6, "close_count": 0, "not_bufo_count": 4, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.78, "cost_per_image_usd": 0.078, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.13, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 51.2}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 32.7}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 35.7}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.7}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-05-sleeping.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 37.1}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.2}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 30.3}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 32.4}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-09-tear.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 47.0}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-pro-conditioned-10-gift.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 39.0}], "counts": {"bufo": 6, "close": 0, "not_bufo": 4, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 39.6}, {"cost_per_image_usd": 0.0388567, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.0388536, 0.0388599], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [0.388536, 0.388599], "id": "google/gemini-2.5-flash-image", "name": "gemini-2.5-flash-image", "run": "2026-09-04-0411-google_gemini-2.5-flash-image-conditioned", "score": 5, "bufo_count": 5, "close_count": 0, "not_bufo_count": 5, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.388567, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.0777134, "cost_per_bufo_range_usd": [0.0777072, 0.0777198], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 47.7}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-02-morning-coffee.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 43.5}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.1}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 50.4}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 50.8}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.5}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.0}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 48.0}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.9}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0411-google_gemini-2.5-flash-image-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 49.7}], "counts": {"bufo": 5, "close": 0, "not_bufo": 5, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 47.5}, {"cost_per_image_usd": 0, "cost_estimated": false, "cost_basis": "Locally hosted", "cost_sample_count": 0, "cost_note": "Configured price: $0 per image for local hosting. Electricity and hardware excluded. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "configuration_note": "Local RTX 3090 Ti; NF4 text encoder, BF16 image model; 1024\u00d71024, 40 steps, seed 42. Locally hosted; $0 per image. Median warm request: 77.3s across nine tasks.", "id": "qwen/qwen-image-2.1-local-nf4-encoder", "name": "qwen-image-2.1-local-nf4-encoder", "run": "2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned", "score": 5, "bufo_count": 5, "close_count": 4, "not_bufo_count": 1, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.0, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Locally hosted", "cost_frontier": true, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-01-meet-proxy.jpg", "rating": "close", "status": "reviewed", "automatic_score": 45.3}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 41.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-03-party-hat.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.9}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-04-thumbs-up.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.5}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-05-sleeping.jpg", "rating": "close", "status": "reviewed", "automatic_score": 43.1}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 44.8}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 32.2}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-08-laptop.jpg", "rating": "close", "status": "reviewed", "automatic_score": 46.1}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.6}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-20-qwen-image-2.1-local-nf4-encoder-conditioned-10-gift.jpg", "rating": "close", "status": "reviewed", "automatic_score": 41.1}], "counts": {"bufo": 5, "close": 4, "not_bufo": 1, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 43.5}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "qwen/qwen-image-3", "name": "qwen-image-3", "run": "2026-09-05-expansion-qwen_qwen-image-3-conditioned", "score": 5, "bufo_count": 5, "close_count": 0, "not_bufo_count": 5, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.33, "cost_per_image_usd": 0.033, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.066, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": "Measured run", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-01-meet-proxy.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.5}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 43.8}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 39.9}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 52.6}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.6}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 49.0}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 31.3}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-08-laptop.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 39.7}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 50.7}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-qwen_qwen-image-3-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 44.2}], "counts": {"bufo": 5, "close": 0, "not_bufo": 5, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 44.1}, {"cost_per_image_usd": 0.03394608333333333, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.0339435, 0.03394875], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [0.33943500000000004, 0.3394875], "id": "google/gemini-3.1-flash-lite-image", "name": "gemini-3.1-flash-lite-image", "run": "2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned", "score": 4, "bufo_count": 4, "close_count": 0, "not_bufo_count": 6, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.33946083333333327, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.08486520833333332, "cost_per_bufo_range_usd": [0.08485875000000001, 0.084871875], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 44.8}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-02-morning-coffee.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 36.2}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.9}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.2}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 37.1}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-06-boba.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.0}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-07-rain.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 40.4}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 48.3}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 45.3}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0217-google_gemini-3.1-flash-lite-image-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 38.9}], "counts": {"bufo": 4, "close": 0, "not_bufo": 6, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 42.6}, {"cost_per_image_usd": 0.21724000000000002, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.21395, 0.22309], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [2.1395, 2.2309], "id": "openai/gpt-5-image", "name": "gpt-5-image", "run": "2026-09-04-0246-openai_gpt-5-image-conditioned", "score": 1, "bufo_count": 1, "close_count": 0, "not_bufo_count": 9, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 2.1724, "cost_observed_attempts": 10, "cost_per_bufo_usd": 2.1724, "cost_per_bufo_range_usd": [2.1395, 2.2309], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 48.4}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-02-morning-coffee.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 49.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 38.2}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 45.9}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 35.8}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 43.7}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 28.8}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 44.0}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 48.0}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0246-openai_gpt-5-image-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.7}], "counts": {"bufo": 1, "close": 0, "not_bufo": 9, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 42.2}, {"cost_per_image_usd": 0.046344166666666665, "cost_estimated": true, "cost_basis": "Est. \u00b7 3 samples", "cost_sample_count": 3, "cost_range_usd": [0.045355, 0.046885], "cost_note": "Estimate from 3 new drawings using the same prompts, reference image, and API route on September 5, 2026. The original bill wasn\u2019t saved. Each sample includes the full request charge. Full ten-attempt cost is estimated from this rate.", "cost_complete": true, "generation_cost_range_usd": [0.45355, 0.46885000000000004], "id": "openai/gpt-5-image-mini", "name": "gpt-5-image-mini", "run": "2026-09-04-0226-openai_gpt-5-image-mini-conditioned", "score": 1, "bufo_count": 1, "close_count": 0, "not_bufo_count": 9, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "demonstrated", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.46344166666666664, "cost_observed_attempts": 10, "cost_per_bufo_usd": 0.46344166666666664, "cost_per_bufo_range_usd": [0.45355, 0.46885000000000004], "cost_per_bufo_basis": "Est. \u00b7 3 samples", "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.2}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-02-morning-coffee.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.2}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 34.6}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 47.1}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 48.8}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.5}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 39.1}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 51.1}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-09-tear.jpg", "rating": "bufo", "status": "reviewed", "automatic_score": 36.9}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-04-0226-openai_gpt-5-image-mini-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 42.1}], "counts": {"bufo": 1, "close": 0, "not_bufo": 9, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 42.9}, {"cost_complete": false, "cost_basis": "Measured returned-image sample", "cost_sample_count": 9, "cost_estimated": false, "cost_note": "Measured charges for 9 returned images; not a full ten-attempt cost.", "id": "black-forest-labs/flux.2-max", "name": "flux.2-max", "run": "2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned", "score": 0, "bufo_count": 0, "close_count": 0, "not_bufo_count": 9, "pending_count": 0, "error_count": 1, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 9, "completed": 9, "incomplete_reason": "1 recorded content-blocked generation failure(s) count as zero points.", "failure_reasons": {"content_blocked": 1}, "observed_ability": "not_demonstrated_in_this_run", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": null, "cost_per_image_usd": 0.1, "cost_observed_attempts": 9, "cost_per_bufo_usd": null, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": null, "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 48.6}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-02-morning-coffee.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 50.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.7}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 50.2}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 49.6}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 41.2}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 49.1}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-09-tear.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 55.5}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-black-forest-labs_flux.2-max-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 54.7}], "counts": {"bufo": 0, "close": 0, "not_bufo": 9, "pending": 0, "error": 1}, "automatic_scored_count": 9, "automatic_score": null}, {"cost_basis": "Measured run", "cost_sample_count": 10, "cost_estimated": false, "cost_note": "API-reported charges for all ten comparable attempts.", "cost_complete": true, "id": "krea/krea-2-large", "name": "krea-2-large", "run": "2026-09-05-expansion-krea_krea-2-large-conditioned", "score": 0, "bufo_count": 0, "close_count": 0, "not_bufo_count": 10, "pending_count": 0, "error_count": 0, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 10, "completed": 10, "incomplete_reason": null, "failure_reasons": {}, "observed_ability": "not_demonstrated_in_this_run", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": 0.6, "cost_per_image_usd": 0.06, "cost_observed_attempts": 10, "cost_per_bufo_usd": null, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": null, "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-01-meet-proxy.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.3}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-02-morning-coffee.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 41.1}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-03-party-hat.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 39.1}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-04-thumbs-up.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 46.0}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-05-sleeping.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 38.4}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-06-boba.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 44.7}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-07-rain.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 40.9}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-08-laptop.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 45.0}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-09-tear.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 45.5}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": "img/2026-09-05-expansion-krea_krea-2-large-conditioned-10-gift.jpg", "rating": "not_bufo", "status": "reviewed", "automatic_score": 43.3}], "counts": {"bufo": 0, "close": 0, "not_bufo": 10, "pending": 0, "error": 0}, "automatic_scored_count": 10, "automatic_score": 43.0}, {"cost_per_image_usd": null, "cost_estimated": true, "cost_basis": "Published token rates", "cost_sample_count": 0, "cost_note": "No images returned. Published Azure rates: $38 per million output-image tokens, $8 per million input-image tokens, and $5 per million input-text tokens. Per-image cost depends on token usage.", "cost_rate_label": "$38 / 1M output tokens + inputs", "cost_source_url": "https://openrouter.ai/api/v1/images/models/microsoft/mai-image-2.6/endpoints", "cost_complete": false, "id": "microsoft/mai-image-2.6", "name": "mai-image-2.6", "run": "2026-09-05-expansion-microsoft_mai-image-2.6-conditioned", "score": 0, "bufo_count": 0, "close_count": 0, "not_bufo_count": 0, "pending_count": 0, "error_count": 10, "unattempted_count": 0, "attempt_count": 10, "attempts_verified": true, "reviewed": 0, "completed": 0, "incomplete_reason": "10 recorded content-blocked generation failure(s) count as zero points.", "failure_reasons": {"content_blocked": 10}, "observed_ability": "not_demonstrated_in_this_run", "observed_ability_note": "Observed in this ten-attempt batch; not a global capability claim.", "generation_cost_usd": null, "cost_observed_attempts": 0, "cost_per_bufo_usd": null, "cost_per_bufo_range_usd": null, "cost_per_bufo_basis": null, "cost_frontier": false, "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "img": null, "rating": null, "status": "recorded_failure", "automatic_score": null}], "counts": {"bufo": 0, "close": 0, "not_bufo": 0, "pending": 0, "error": 10}, "automatic_scored_count": 0, "automatic_score": null}], "tasks": [{"id": "01-meet-proxy", "title": "Bufo proxying between two bots", "prompt": "Bufo happily proxying: the frog holds a horizontal yellow double-headed arrow bar, one light-blue robot head in a speech bubble on each end. All three elements in one row, centered, emoji-size readability.", "checks": ["yellow double-arrow bar present and horizontal", "two light-blue robot heads in speech bubbles, one per side", "bufo is the middle figure holding the bar"]}, {"id": "02-morning-coffee", "title": "Bufo holding coffee", "prompt": "Bufo holding a small white coffee mug with both hands, looking at it with tired deadpan eyes. A few small steam curls above the mug.", "checks": ["white mug held in both hands", "steam curls", "tired deadpan expression, no smile"]}, {"id": "03-party-hat", "title": "Bufo with party hat", "prompt": "Bufo wearing a small striped party hat tilted on its head, sitting and staring blankly forward. The hat is the only festive element; the frog is unimpressed.", "checks": ["striped party hat, tilted", "seated full-body or bust composition", "blank unimpressed expression"]}, {"id": "04-thumbs-up", "title": "Bufo giving thumbs-up", "prompt": "Bufo giving a thumbs-up with one stubby arm, facing the viewer, slight awkward smirk at most.", "checks": ["single thumbs-up gesture readable at 64px", "faces viewer", "no open-mouth smile"]}, {"id": "05-sleeping", "title": "Bufo asleep", "prompt": "Bufo asleep: slumped blob posture, eyes closed as two short curved lines, a small 'Z' above its head.", "checks": ["closed eyes as simple lines (no detailed eyelids)", "slumped/resting posture", "'Z' present"]}, {"id": "06-boba", "title": "Bufo holding boba tea", "prompt": "Bufo holding a cup of boba milk tea with both hands, fat straw sticking out, dark tapioca pearls visible at the bottom of the cup. Blank unimpressed stare at the viewer.", "checks": ["cup with visible boba pearls and a straw", "held in both hands", "blank expression, no smile"]}, {"id": "07-rain", "title": "Bufo under a rain cloud", "prompt": "Bufo sitting under a small grey rain cloud, neat raindrops falling around it. Bufo looks upward at the cloud with worried sidelong eyes.", "checks": ["small grey cloud positioned above bufo", "visible raindrops", "upward worried gaze"]}, {"id": "08-laptop", "title": "Bufo at a laptop", "prompt": "Bufo seated facing a tiny laptop computer, screen glow lighting its face from below. Deadpan tired stare at the screen, late-night coding energy.", "checks": ["laptop present and faced by bufo", "screen glow on face", "tired deadpan stare, no smile"]}, {"id": "09-tear", "title": "Bufo shedding one tear", "prompt": "Bufo with a single tear drop rolling down its cheek, worried sidelong glance intact. Sad but composed \u2014 one tear only, no sobbing, no mouth.", "checks": ["exactly one tear drop", "sidelong worried glance", "no mouth of any kind"]}, {"id": "10-gift", "title": "Bufo offering a gift", "prompt": "Bufo holding out a small wrapped gift box with a bow toward the viewer, arms extended. Blank unimpressed stare \u2014 the gift is offered out of obligation, not joy.", "checks": ["wrapped gift box with a bow", "held out toward viewer", "blank unimpressed stare"]}], "reference_img": "img/froge-original.jpg", "method": {"version": "reviewed-comparison-v1", "label": "Bufo count", "scale": "0\u201310, recognizable human Bufo labels over ten fixed attempts", "denominator": 10, "source_batch": "reviewed-comparison-v1-96f165e976d9cce5", "description": "A model earns one point only for an explicit human Bufo label. Close and Not Bufo earn zero. Every row uses ten fixed attempts; verified generation failures contribute zero points, while unattempted or unreviewed attempts prevent ranking.", "policy": "restored human recognizable-Bufo comparison; exact-hash joins with recorded precedence; no classifier fallback; no partial credit; visitor votes never change rows", "cost_policy": "Cost per accepted Bufo uses a complete comparable ten-attempt recorded or estimated cost divided by Bufo count. Partial costs and zero-pass rows remain undefined.", "label_coverage": {"bufo": 97, "close": 4, "conflicts": 7, "expected": 179, "not_bufo": 78, "records": 179}, "observed_ability": "A positive count demonstrates ability in this batch only; zero is not a global inability claim."}, "scoring": {"method": "human recognizable Bufo count", "version": "reviewed-comparison-v1", "classifier_status": "not used"}, "frontier": [{"model": "qwen/qwen-image-2.1-local-nf4-encoder", "short": "qwen-image-2.1-local-nf4-encoder", "cost": 0, "score": 5}, {"model": "meta/muse-image", "short": "muse-image", "cost": 0.01, "score": 9}], "value_pts": [{"short": "muse-image", "model": "meta/muse-image", "cost": 0.01, "score": 9, "frontier": true}, {"short": "gpt-5.4-image-2", "model": "openai/gpt-5.4-image-2", "cost": 0.231827, "score": 9, "frontier": false}, {"short": "gpt-image-2", "model": "openai/gpt-image-2", "cost": 0.012791, "score": 9, "frontier": false}, {"short": "recraft-v4-styles-pro", "model": "recraft/recraft-v4-styles-pro", "cost": 0.10500000000000001, "score": 8, "frontier": false}, {"short": "grok-imagine-image-2.0", "model": "x-ai/grok-imagine-image-2.0", "cost": 0.07, "score": 8, "frontier": false}, {"short": "gemini-3-pro-image", "model": "google/gemini-3-pro-image", "cost": 0.13930066666666666, "score": 7, "frontier": false}, {"short": "gemini-3.1-flash-image", "model": "google/gemini-3.1-flash-image", "cost": 0.06789216666666666, "score": 7, "frontier": false}, {"short": "riverflow-v2.5-pro", "model": "sourceful/riverflow-v2.5-pro", "cost": 0.2598604, "score": 7, "frontier": false}, {"short": "seedream-5-0-pro", "model": "bytedance-seed/seedream-5-0-pro", "cost": 0.09, "score": 6, "frontier": false}, {"short": "qwen-image-3-pro", "model": "qwen/qwen-image-3-pro", "cost": 0.078, "score": 6, "frontier": false}, {"short": "gemini-2.5-flash-image", "model": "google/gemini-2.5-flash-image", "cost": 0.0388567, "score": 5, "frontier": false}, {"short": "qwen-image-2.1-local-nf4-encoder", "model": "qwen/qwen-image-2.1-local-nf4-encoder", "cost": 0, "score": 5, "frontier": true}, {"short": "qwen-image-3", "model": "qwen/qwen-image-3", "cost": 0.033, "score": 5, "frontier": false}, {"short": "gemini-3.1-flash-lite-image", "model": "google/gemini-3.1-flash-lite-image", "cost": 0.03394608333333333, "score": 4, "frontier": false}, {"short": "gpt-5-image", "model": "openai/gpt-5-image", "cost": 0.21724000000000002, "score": 1, "frontier": false}, {"short": "gpt-5-image-mini", "model": "openai/gpt-5-image-mini", "cost": 0.046344166666666665, "score": 1, "frontier": false}, {"short": "krea-2-large", "model": "krea/krea-2-large", "cost": 0.06, "score": 0, "frontier": false}], "factoids": {}, "automatic_judging": {"version": "bufo-score-v1", "status": "experimental", "label": "Automatic Bufo likeness", "head_sha256": "800bc15b9d8259927fa7e94557c84664c82f0ca5f74e99435dbe6c95c21de2ba", "config_sha256": "72b3e8ea4105c7726e37352f50821fdcdeaf371d870c25ac51ad1963f5f7b0b6", "scale": [0, 100], "description": "Frozen SigLIP 2 classifier; mean of ten image scores. Visual likeness only, not a human pass rate or a task-completion score. The automatic judge still makes mistakes; human labels remain separate."}}; | |