Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
142 Open 899 Closed
142 Open 899 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 Conv->Relu->Concat Q/DQ insertion gap
#1398 opened May 6, 2026 by willg-nv Contributor Loading…
8
[Quantization] Saturate NVFP4 export FP8 scale cast to avoid NaN cherry-pick-0.44.0 After code freeze, cherry-pick to release branch for next rc (bulk update). Only for bug fixes / doc
#1397 opened May 6, 2026 by cjluo-nv Collaborator Loading…
2 tasks done
7
[Recipes][LLM PTQ] Add nvfp4 MSE+FP8-cast-KV recipes (experts_only / mlp_only) + --recipe in example scripts
#1391 opened May 4, 2026 by cjluo-nv Collaborator Loading…
2 of 4 tasks
2
[Quantization] Fused Triton kernel for NVFP4 FP8 scale sweep search
#1387 opened May 4, 2026 by cjluo-nv Collaborator Loading…
3 tasks done
23
Add unit test for checking any leak of temporary augmented onnx files, on exception during ONNX INT4 AWQ quantization
#1383 opened May 3, 2026 by vishalpandya1990 Contributor Loading…
2
fixes for fused moe (qwen3.6, GLM5.1 + MSE calibration
#1382 opened May 2, 2026 by Fridah-nv Contributor Loading…
29
AutoQuant for VLM
#1381 opened May 1, 2026 by meenchen Contributor • Draft
4
feat(launcher): add DFlash support for DeepSeek-V4-Flash target model
#1379 opened Apr 30, 2026 by ChenhanYu Collaborator Loading…
7
Use trtexec_safe on safety platforms when using remoteAutoTuning
#1378 opened Apr 30, 2026 by dthienan-nv Contributor Loading…
26
Add Nemotron-3-Nano-30B-A3B-BF16 e2e tutorial: Prune + Distill + Quantize + Nemo Evaluator + vLLM deployment
#1376 opened Apr 30, 2026 by kevalmorabia97 Collaborator • Draft
4
k25 dflash hardcode support
#1367 opened Apr 29, 2026 by h-guo18 Contributor • Draft
2
Experiment: MXFP4 -> NVFP4 conversion MSE study (scratch)
#1364 opened Apr 28, 2026 by cjluo-nv Collaborator • Draft
3 tasks
3
Support Mixed precision & Static MSE PTQ in MCore export; Nemotron Super v3 NVFP4 recipe
#1363 opened Apr 28, 2026 by jenchen13 Contributor Loading…
32
[SKILL.md Chore] make .agents/ the cannonical agent-skills location
#1362 opened Apr 28, 2026 by shljessie Loading…
22
Enable runtime optimization
#1358 opened Apr 28, 2026 by grzegorz-k-karch Contributor • Draft
@grzegorz-k-karch
4
Add pre-built evaluation recipes for common benchmarks
#1357 opened Apr 27, 2026 by kaix-nv Contributor Loading…
26
[OMNIML-4021]: align local JSONL loading with HF datasets path + keep original behaviour
#1345 opened Apr 24, 2026 by shengliangxu Collaborator Loading…
3 tasks done
15
[minor] fixes for layerwise calib + MSE
#1344 opened Apr 24, 2026 by Fridah-nv Contributor Loading…
@Fridah-nv
9
DSV4 dequant on the fly
#1341 opened Apr 24, 2026 by mxinO Contributor • Draft
4
Update
#1338 opened Apr 23, 2026 by jingyu-ml Contributor • Draft
@jingyu-ml
3
[OMNIML-3934] Guidelines and precommit hook for pydantic backward compatbility
#1333 opened Apr 23, 2026 by jenchen13 Contributor Loading…
8
[Refactor] speculative decoding: use mto config subsystem
#1328 opened Apr 23, 2026 by h-guo18 Contributor Loading…
@h-guo18
23
Quantize lm_head + embedding for Nemotron-H, add NVFP4 W4A16 recipe
#1327 opened Apr 22, 2026 by ajrasane Contributor Loading…
3 of 5 tasks
@ajrasane
14
Update the DMD2 at the first stage
#1326 opened Apr 22, 2026 by jingyu-ml Contributor • Draft
@jingyu-ml
13
Fix NVFP4 quantization for Qwen3.x MoE models (4 silent-failure bugs)
#1323 opened Apr 22, 2026 by erictinkeredapps Loading…
9
ProTip! Exclude everything labeled bug with -label:bug.