What happened: The initial upload (2026-04-15) used ignore=[«lmhead»] in the llm-compressor recipe, which meant the 62 MoE routers (blocksparsemoe.gate) got quantized along with the expert weights. vLLM’s MiniMax-M2 loader expects an unquantized ReplicatedLinear` router and fails at engine-init with: This is a hard load failure — the engine never initializes, so no tokens are generated. (The earlier «degraded output» framing understated the severity.) Root cause: Missing MoE-aware entries in the llm-compressor ignore list. The correct pattern (per saricles/MiniMax-M2.5-REAP-139B-A10B-NVFP4-GB10): Fix: This variant was re-rolled 2026-04-16 with the corrected recipe. quantizationconfig.ignore now lists all 62 per-layer router gates alongside lmhead. Verification: config.json on this repo now contains 62 model.layers.N.blocksparsemoe.gate entries in the ignore list. Loaders should open the model without the KeyError above. Credit: Thanks to the community user who reported this first on the NVFP4-GB10 DGX Spark load. The saricles reference repo was invaluable for confirming the exact pattern. Unaffected variants (no re-roll needed): BF16 safetensors, all GGUF quantizations. This…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: dervig
Теги: minimax_m2, minimax, moe, reap, nvfp4, w4a4, fp4, blackwell
Лайков: 4 | Загрузок: 102
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
