-
Repoint build instructions to charlie12345/ROCmFPX (ROCmFPX FP3/4/6/8 repo)
-
Upload README.md with huggingface_hub
-
consolidate to single best-balance rocmfp4 quant
-
consolidate to single best-balance rocmfp4 quant
-
Add best-of-both recipe
-
Card: bundled mmproj-F32.gguf for vision
-
Bundle Qwen3-VL mmproj (vision projector, projection_dim 5120)
-
Add best-of-both recipe (f16 emb + Q6 head + all-dual body)
-
Card: add vision flags to the start command (--mmproj + --image-min-tokens 1024)
-
Card: best-of-both recommended, KL fidelity findings, BCB result, vision flags
-
Add Qwen ASCII logo header (centered logo+title, full-width spec table)
-
Correct mmproj/cache claim: source says only --cache-reuse is disabled w/ multimodal (not 'cache reuse entirely'); rest unmeasured. (27B: + upstream-benchmark pointer)
-
Add upstream SWE-bench (Jackrong's base Qwopus-Coder, 67.0% thinking-off Q5_K_M) — clearly labeled NOT re-measured on this quant
-
f16 KV: state config only, no acceptance claim either way (pending more rigorous testing)
-
Correct unsupported claim: f16 KV is higher fidelity, NOT a measured acceptance lever (short-context A/B showed no difference vs q4_0)
-
Card: square corners on all tables (overflow:hidden + border-radius:0) + red critical-fork callout + amber section numbers
-
Card: datasheet/neo-brutalist redesign — monospace, hard square borders, numbered sections, symmetric spec grid, theme-adaptive (currentColor)
-
Card: styled HTML redesign (gradient header, pill badges, styled tables, GitHub alerts) — no collapsibles, theme-adaptive body, Jackrong-inspired
-
Card: vertical copy-paste launch command (one flag/line, \-continued, syntax-verified) + collapsible 'what every flag does' reference
-
Card redesign: centered header + badge bar, at-a-glance/files tables, collapsible launch/build/OpenCode, vision + perf sections (prototype for fleet-wide structure)
-
Card: add Vision section (--mmproj + --image-min-tokens 1024) and bundled chat_template.jinja note
-
Add froggeric v20 fixed chat template (vision + inline think-toggle + tool anti-stall); use via --chat-template-file
-
Card: trim root-cause detail; document thinking-off cache fix (preserve_thinking) + OpenCode fork setup
-
Card: document agentic tool-calling fix (native tool_call:true), thinking on/off cache trade-off, update launch flags (n-max 5, cache-ram 64G)
-
provenance: set base_model_relation=quantized
-
Re-quant from BF16: genuine f16 emb, Q6_K head (replaces fake-f16)
-
Re-quant from BF16: genuine f16 embeddings (replaces Q8-sourced fake-f16)
-
Card: re-quantized from BF16 (genuine f16 emb); no imatrix (measured mixed on this coder)
-
Card: honest note on the f16 embeddings (Q8 source -> f16-typed but Q8 precision)
-
Add model card (two variants, off-thinking note, credits)
-
Add Q6-head variant
-
Add base variant (ROCmFP4 STRIX + f16 emb, no imatrix; from Q8_0)
-
initial commit
