unsloth/unsloth/models
Daniel Han dde18ddd28 Fix FP8 MoE loader: guard import, pass auth to FastModel
1. Wrap moe_utils_fp8 import in try/except so unsloth does not
   crash when paired with an older unsloth_zoo that lacks the
   module. Falls back to a no-op stub.

2. Pass token and trust_remote_code to get_model_name() and
   _has_prequantized_fp8_config() in the FastModel path, matching
   the FastLanguageModel path. Without this, private/gated repos
   fail silently and fall through to incorrect offline quantization.
2026-03-16 14:32:34 +00:00
..
__init__.py Revert "feat: Add Mixtral model support" 2026-03-13 22:38:49 -07:00
_utils.py [pre-commit.ci] auto fixes from pre-commit.com hooks 2026-03-16 07:10:49 +00:00
cohere.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
dpo.py Formatting & bug fixes (#3563) 2025-11-07 06:00:22 -08:00
falcon_h1.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
gemma.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
gemma2.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
glm4_moe.py [WIP] cleanup 2026-03-15 14:49:07 +00:00
granite.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
llama.py Fixup mapper issues and resolve properly (#4124) 2026-03-03 06:57:25 -08:00
llama4.py Qwen 3 2025-05-02 03:09:44 -07:00
loader.py Fix FP8 MoE loader: guard import, pass auth to FastModel 2026-03-16 14:32:34 +00:00
loader_utils.py [pre-commit.ci] auto fixes from pre-commit.com hooks 2026-03-16 07:10:49 +00:00
mapper.py Fixup mapper issues and resolve properly (#4124) 2026-03-03 06:57:25 -08:00
mistral.py Fix/pr 3699 leftpad prefill main (#4100) 2026-02-25 07:21:04 -08:00
qwen2.py Revert "[pre-commit.ci] auto fixes from pre-commit.com hooks" 2025-12-01 07:24:58 -08:00
qwen3.py Undo qwen3 rope changes 2026-03-16 07:10:10 +00:00
qwen3_moe.py Fix correctness bugs across multiple model files (#3813) 2026-01-01 02:36:33 -08:00
rl.py Revert "patch vlm trainer to resize images" 2026-03-13 22:39:07 -07:00
rl_replacements.py [trl] Trl v0.28 (and above) rl fixes (#4156) 2026-03-09 12:01:49 -04:00
sentence_transformer.py Fix EmbeddingGemma float16 NaN via FORCE_FLOAT32 for gemma3_text (#4014) 2026-02-10 01:40:13 -08:00
vision.py Fixup mapper issues and resolve properly (#4124) 2026-03-03 06:57:25 -08:00