Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
51 Open 337 Closed
51 Open 337 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

[5693592][ONNX-customOp][Autocast] Fix QuantizeLinear node output type
#671 opened Dec 9, 2025 by gcunhase Loading…
1
Fix a bug in IterativeChannelContributionHook + tools for activation hooks analysis
#670 opened Dec 9, 2025 by danielkorzekwa Loading…
1
Support KIMI K2 Thinking int4 checkpoint PTQ
#669 opened Dec 9, 2025 by cjluo-nv Loading…
12
Refactor: Eagle data loading
#668 opened Dec 8, 2025 by h-guo18 • Draft
1
Remove runtime.py and directly use torch dist utils + remove unused functions
#667 opened Dec 8, 2025 by kevalmorabia97 Loading…
9
Add Quantizers for Qwen3VLMoeTextDecoderLayer
#666 opened Dec 8, 2025 by soodoshll Loading…
4
Refactor and clean up hf_ptq.py
#665 opened Dec 8, 2025 by shengliangxu Loading…
1
Yeyu/hf eagle medusa
#664 opened Dec 8, 2025 by yeyu-nvidia Loading…
@yeyu-nvidia
3
Replace mip package with pulp
#663 opened Dec 8, 2025 by kevalmorabia97 Loading…
us gesvd as solver
#661 opened Dec 8, 2025 by andompesta Loading…
2
[5336829][AutoCast] Support subgraphs
#659 opened Dec 7, 2025 by galagam Loading…
2
config file based modelopt config 1/N
#657 opened Dec 6, 2025 by shengliangxu • Draft
2
[OMNIML-2244] Add E2E example for mixed precision quantization and ONNX export
#656 opened Dec 5, 2025 by ajrasane Loading…
13
Kimi-k2 calib+export
#655 opened Dec 5, 2025 by jingyu-ml Loading…
11
Support model export for int4 wo
#653 opened Dec 5, 2025 by meenchen Loading…
@meenchen
1
[5676209][ONNX][Autocast] Add check for input bs vs calibration data bs
#652 opened Dec 5, 2025 by gcunhase Loading…
3
[NVBUG: 5701937]Clear GPU cache for 3D weight tensors
#649 opened Dec 4, 2025 by cjluo-nv Loading…
1
Added support for TElinear ops
#632 opened Dec 2, 2025 by kinjalpatel27 Loading…
18
Address security concerns in code
#626 opened Dec 2, 2025 by kevalmorabia97 Loading…
2 of 3 tasks
5
refactor: eagle config rewrite; update doc
#624 opened Dec 2, 2025 by h-guo18 Loading…
@h-guo18
2
Updated dependencies in Windows examples
#622 opened Dec 1, 2025 by hthadicherla Loading…
7
Optimize calibrate_draft_vocab to read only required lines when calib…
#618 opened Nov 27, 2025 by Ofir408 Loading…
3
MLA eagle for K2
#615 opened Nov 27, 2025 by h-guo18 • Draft
1
draft: Add per block MSE for NVFP4 and INT4
#613 opened Nov 27, 2025 by Fridah-nv • Draft
@Fridah-nv
4
Convert compressed-tensor int4 format to GPTQ int4 format
#590 opened Nov 20, 2025 by Edwardf0t1 Loading…
2
ProTip! Add no:assignee to see everything that’s not assigned.