Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
96 Open 507 Closed
96 Open 507 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix: quant config error on quantized offline eagle
#925 opened Feb 24, 2026 by h-guo18 Loading…
6
Support decoder block-level sequential calibration
#924 opened Feb 24, 2026 by sugunav14 Loading…
8
Fix skip softmax defaults
#923 opened Feb 24, 2026 by rohansjoshi Loading…
2
Enable multinode training for HF speculative decoding
#922 opened Feb 23, 2026 by yeyu-nvidia Loading…
7
Diffusers 2:4 Sparse Attention
#921 opened Feb 23, 2026 by jingyu-ml • Draft
2
example demonstrating how to train CosmosReason2 Eagle3
#920 opened Feb 23, 2026 by skierat Loading…
5
Add Qwen3-VL support to Minitron pruning
#919 opened Feb 23, 2026 by eagle705 • Draft
1
disable rope scaling for training, add yarn during export
#917 opened Feb 23, 2026 by h-guo18 • Draft
8
Add 2:4 Sparse Attention
#916 opened Feb 22, 2026 by kaix-nv • Draft
9
Feat: Speculatice Decoding export with quantization support
#913 opened Feb 21, 2026 by h-guo18 Loading…
12
Add support for export ComfyUI compatible checkpoint for diffusion model(e.g., LTX-2)
#911 opened Feb 20, 2026 by ynankani Loading…
13
Support force tokens to % of total experts during calibration
#910 opened Feb 20, 2026 by cjluo-nv Loading…
41
fix
#908 opened Feb 19, 2026 by h-guo18 • Draft
3
PTQ and QAD with Qwen Image
#905 opened Feb 18, 2026 by AliesTaha Loading…
14
Support mbridge distillation for any_model
#904 opened Feb 18, 2026 by danielkorzekwa Loading…
24
Enable Qwen3.5-MoE PTQ
#897 opened Feb 16, 2026 by Edwardf0t1 • Draft
4
Fix DeepSpeed import crash on runtime-only CUDA and improve NVFP4 uncalibrated weight error
#896 opened Feb 16, 2026 by debo3 Loading…
2
Add Qwen3VL
#895 opened Feb 16, 2026 by hychiang-git Loading…
2
gpt-oss 20b support
#889 opened Feb 13, 2026 by chochowski Loading…
10
Implicit Gemm NVFP4 on Conv3D
#886 opened Feb 13, 2026 by jingyu-ml Loading…
@jingyu-ml
12
Add support for offline speculative decoding model PTQ
#883 opened Feb 12, 2026 by yeyu-nvidia • Draft
3
update qwen quant
#880 opened Feb 11, 2026 by zhewenl • Draft
5
Update README.md for DMS (fix cd experimental/DMS to cd Model-Optimizer/experimental/DMS)
#879 opened Feb 10, 2026 by faridlazuarda Loading…
4
[OMNIML-2914] Fix export of fused layernorm weights for TE spec
#876 opened Feb 10, 2026 by yueshen2016 Loading…
4
SpecDec Bench: February Update
#875 opened Feb 10, 2026 by IzzyPutterman Loading…
31
ProTip! Follow long discussions with comments:>50.