{ "passed": true, "method": "Clean wheel installation served from a directory outside the source checkout; temporary API reused existing pinned native GPU server; API stopped afterward.", "version": "0.1.1", "source_matches_benchmark": true, "source_sha256": { "__init__.py": "b6d87a4431c3d0de11cc824d2e3f2e6b4e267fe3c2aa07bf6b39098b85771d47", "app.py": "949cb10750883a7b0ff80f58c856151d7ce2390a745342b40037461a2159c153", "core.py": "1b8b4cce7260bcfbd4f8d8d0daf8ebfacabc9c202911f652674691fc30c8a1fd", "models.py": "8a61790f307bc21decc4c768d853f34dca5edc0ba2f5c9c585a77339396338d3" }, "examples": { "support-request.json": { "model": "bev-bonsai-27b", "backend": "prism_llama_cpp", "backend_build": "b1-9a9394a", "mode": "llama_full_vocab_categorical_scoring", "strategy": "batch", "requested_strategy": "auto", "strategy_implementation": "bounded_concurrent_native_requests", "parsed_json": { "route": "billing" }, "fields": { "route": { "value": "billing", "selected_value": "billing", "selected_label": "A", "selected_token_id": 32, "probability": 0.9989061260210375, "abstained": false, "candidate_probability_mass": 0.9086667884355636, "candidates": [ { "value": "billing", "label": "A", "token_id": 32, "probability": 0.9989061260210375, "logprob": -0.09687129408121109 }, { "value": "technical", "label": "B", "token_id": 33, "probability": 0.0008516050271616897, "logprob": -7.1641645431518555 }, { "value": "unknown", "label": "C", "token_id": 34, "probability": 0.0002422689518008115, "logprob": -8.421238899230957 } ], "elapsed_ms": 408.331, "queue_ms": 0.003, "backend_requests": 1, "score_retrieval": "selected_token_logprobs", "requested_top_n": 3, "prompt_tokens": 142, "backend_timings": { "cache_n": 0, "prompt_n": 142, "prompt_ms": 284.805, "prompt_per_token_ms": 2.005669014084507, "prompt_per_second": 498.58675233931984, "predicted_n": 1, "predicted_ms": 0.001, "predicted_per_token_ms": 0.0, "predicted_per_second": 0.0 }, "usage": { "prompt_tokens": 142, "completion_tokens": 1, "total_tokens": 143 } } }, "scores_are_calibrated": false, "score_source": "pre_sampling_full_vocabulary_logprobs", "selected_token_logprobs": true, "score_temperature": 1.0, "abstained_fields": [], "elapsed_ms": 412.751, "prompt_preparation_ms": 4.37, "backend_requests": 1, "scored_fields": 1, "max_prompt_tokens": 142, "max_model_len": 16384, "backend_slots": 2, "cache_block_tokens": null, "usage": { "prompt_tokens": 142, "completion_tokens": 1, "total_tokens": 143 } }, "systemone-request.json": { "model": "bev-bonsai-27b", "answers": { "queue": { "type": "choice", "choice": "billing", "probabilities": { "billing": 0.9981141129355442, "technical": 0.0017793923442547745, "general": 0.00010649472020099965 }, "confidence": 0.9981141129355442 }, "outage": { "type": "noul", "noul": 0.002480204845362905 }, "priority": { "type": "score", "score": 1.0002383132357986, "probabilities": { "0": 0.0007058337315065528, "1": 0.9983500193011885, "2": 0.0009441469673050024 }, "legend": { "0": "General question", "1": "Billing issue", "2": "Outage" }, "confidence": 0.9983500193011885 } }, "usage": { "input_tokens": 547, "output_tokens": 3, "prompt_tokens": 547, "completion_tokens": 3, "total_tokens": 550 }, "usage_accounting": "logical_prompt_and_scoring_tokens_including_retries", "scores_are_calibrated": false, "confidence_definition": "maximum_candidate_relative_probability", "evaluation_trace": { "model": "bev-bonsai-27b", "backend": "prism_llama_cpp", "backend_build": "b1-9a9394a", "mode": "llama_full_vocab_categorical_scoring", "strategy": "batch", "requested_strategy": "auto", "strategy_implementation": "bounded_concurrent_native_requests", "parsed_json": { "queue": "billing", "outage": false, "priority": "1" }, "fields": { "queue": { "value": "billing", "selected_value": "billing", "selected_label": "A", "selected_token_id": 32, "probability": 0.9981141129355442, "abstained": false, "candidate_probability_mass": 0.8587864055938211, "candidates": [ { "value": "billing", "label": "A", "token_id": 32, "probability": 0.9981141129355442, "logprob": -0.1541227102279663 }, { "value": "technical", "label": "B", "token_id": 33, "probability": 0.0017793923442547745, "logprob": -6.483718395233154 }, { "value": "general", "label": "C", "token_id": 34, "probability": 0.00010649472020099965, "logprob": -9.299650192260742 } ], "elapsed_ms": 668.761, "queue_ms": 0.002, "backend_requests": 1, "score_retrieval": "selected_token_logprobs", "requested_top_n": 3, "prompt_tokens": 190, "backend_timings": { "cache_n": 0, "prompt_n": 190, "prompt_ms": 580.518, "prompt_per_token_ms": 3.0553578947368423, "prompt_per_second": 327.2938995862316, "predicted_n": 1, "predicted_ms": 0.001, "predicted_per_token_ms": 0.0, "predicted_per_second": 0.0 }, "usage": { "prompt_tokens": 190, "completion_tokens": 1, "total_tokens": 191 } }, "outage": { "value": false, "selected_value": false, "selected_label": "B", "selected_token_id": 33, "probability": 0.9975197951546371, "abstained": false, "candidate_probability_mass": 0.9867459180461733, "candidates": [ { "value": true, "label": "A", "token_id": 32, "probability": 0.002480204845362905, "logprob": -6.012756824493408 }, { "value": false, "label": "B", "token_id": 33, "probability": 0.9975197951546371, "logprob": -0.015825986862182617 } ], "elapsed_ms": 666.695, "queue_ms": 0.001, "backend_requests": 1, "score_retrieval": "selected_token_logprobs", "requested_top_n": 2, "prompt_tokens": 164, "backend_timings": { "cache_n": 0, "prompt_n": 164, "prompt_ms": 583.47, "prompt_per_token_ms": 3.5577439024390247, "prompt_per_second": 281.07700481601455, "predicted_n": 1, "predicted_ms": 0.001, "predicted_per_token_ms": 0.0, "predicted_per_second": 0.0 }, "usage": { "prompt_tokens": 164, "completion_tokens": 1, "total_tokens": 165 } }, "priority": { "value": "1", "selected_value": "1", "selected_label": "B", "selected_token_id": 33, "probability": 0.9983500193011885, "abstained": false, "candidate_probability_mass": 0.9648334149708054, "candidates": [ { "value": "0", "label": "A", "token_id": 32, "probability": 0.0007058337315065528, "logprob": -7.291930675506592 }, { "value": "1", "label": "B", "token_id": 33, "probability": 0.9983500193011885, "logprob": -0.037451162934303284 }, { "value": "2", "label": "C", "token_id": 34, "probability": 0.0009441469673050024, "logprob": -7.001028537750244 } ], "elapsed_ms": 1076.032, "queue_ms": 666.574, "backend_requests": 1, "score_retrieval": "selected_token_logprobs", "requested_top_n": 3, "prompt_tokens": 193, "backend_timings": { "cache_n": 0, "prompt_n": 193, "prompt_ms": 322.763, "prompt_per_token_ms": 1.6723471502590672, "prompt_per_second": 597.9619720971734, "predicted_n": 1, "predicted_ms": 0.001, "predicted_per_token_ms": 0.0, "predicted_per_second": 0.0 }, "usage": { "prompt_tokens": 193, "completion_tokens": 1, "total_tokens": 194 } } }, "scores_are_calibrated": false, "score_source": "pre_sampling_full_vocabulary_logprobs", "selected_token_logprobs": true, "score_temperature": 1.0, "abstained_fields": [], "elapsed_ms": 1083.211, "prompt_preparation_ms": 6.793, "backend_requests": 3, "scored_fields": 3, "max_prompt_tokens": 193, "max_model_len": 16384, "backend_slots": 2, "cache_block_tokens": null, "usage": { "prompt_tokens": 547, "completion_tokens": 3, "total_tokens": 550 } } } } }