3 people 960d628f46 mamba2: remove hardcoded 2x expansion factor and invalid d_inner % d_state check (#23082) ...

* mamba2: remove hardcoded 2x expansion factor, support any expand value

* mamba2: remove invalid d_inner %% d_state check (unrelated parameters)

* Update convert_hf_to_gguf.py: make expand optional with default 2

Co-authored-by: Sigbjørn Skjæret <[email protected]>

* mamba2: apply expand fix to refactored conversion/mamba.py

* also check for mamba_expand

---------

Co-authored-by: Sigbjørn Skjæret <[email protected]>
Co-authored-by: Sigbjørn Skjæret <[email protected]>

2026-06-26 08:50:54 +03:00

..

__init__.py

docs : Eagle3 qwen3 draft model support (#24977)

2026-06-25 15:58:00 +03:00

afmoe.py

…

arctic.py

…

baichuan.py

…

bailingmoe.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

base.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

bert.py

model : support granite multilingual embeddings R2 (ibm-granite/granite-embedding-{97,311}m-multilingual-r2) (#22716)

2026-06-02 17:55:11 +02:00

bitnet.py

…

bloom.py

…

chameleon.py

…

chatglm.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

codeshell.py

…

cogvlm.py

…

command_r.py

Add arch support for cohere2-MoE (#24260)

2026-06-13 19:49:00 +02:00

dbrx.py

…

deci.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

deepseek.py

mtmd: model: unlimited-ocr: converter + parity test (#24969)

2026-06-24 18:20:22 +02:00

dots1.py

…

dotsocr.py

…

dream.py

…

ernie.py

…

exaone.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

falcon_h1.py

…

falcon.py

…

gemma.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

glm.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

gpt2.py

…

gpt_oss.py

…

gptneox.py

…

granite.py

model: Granite Speech Plus (#24818)

2026-06-23 12:03:31 +02:00

grok.py

…

grovemoe.py

…

hunyuan.py

mtmd, model : merge HunyuanOCR into HunyuanVL and fix OCR vision precision (#23329)

2026-05-21 00:35:37 +02:00

internlm.py

…

internvl.py

…

jais.py

…

jamba.py

…

januspro.py

…

kimi_linear.py

…

kimivl.py

…

lfm2.py

model : Add LFM2.5-ColBERT-350M and LFM2.5-Embedding-350M (#24913)

2026-06-24 09:49:46 +03:00

lighton_ocr.py

…

llada.py

…

llama4.py

…

llama.py

docs : Eagle3 qwen3 draft model support (#24977)

2026-06-25 15:58:00 +03:00

llava.py

…

maincoder.py

…

mamba.py

mamba2: remove hardcoded 2x expansion factor and invalid d_inner % d_state check (#23082)

2026-06-26 08:50:54 +03:00

mellum.py

model: add Mellum architecture (#23966)

2026-06-02 22:11:12 +03:00

mimo.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

minicpm.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

minimax.py

…

mistral3.py

…

mistral.py

convert : fix conversion for Mistral-Medium-3.5-128B (#24268)

2026-06-07 21:41:39 +02:00

mpt.py

…

nemotron.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

olmo.py

…

openelm.py

…

orion.py

…

pangu.py

…

phi.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

pixtral.py

…

plamo.py

…

plm.py

…

qwen3vl.py

…

qwen.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

qwenvl.py

…

refact.py

…

rwkv.py

…

sarashina2.py

…

smallthinker.py

…

smolvlm.py

…

stablelm.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

starcoder.py

…

step3.py

convert : more consistent handling of rope_parameters (#24833)

2026-06-20 13:42:36 +03:00

t5.py

…

talkie.py

model : add support for talkie-1930-13b (#22596)

2026-05-26 07:57:38 +03:00

ultravox.py

…

wavtokenizer.py

…

xverse.py

…

youtuvl.py

…