Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
147 Open 911 Closed
147 Open 911 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

DLA surgeries added as part of ModelOpt
#1415 opened May 8, 2026 by mgohil-png Contributor Loading…
3
[OMNIML-4312] training_support
#1414 opened May 8, 2026 by ChenhanYu Collaborator • Draft
2
fix(quantization): enforce Blackwell MX-unit alignment for NVFP4 block-size validation
#1413 opened May 7, 2026 by makroumi Loading…
6
[2/2Refactor]: Spec-dec recipe lib: composable + per-model recipe
#1412 opened May 7, 2026 by h-guo18 Contributor • Draft
4
[6106576] Restore qdq_utils symbols used by edgellm 0.6.1 cherry-pick-0.44.0 After code freeze, cherry-pick to release branch for next rc (bulk update). Only for bug fixes / doc
#1408 opened May 7, 2026 by ajrasane Contributor Loading…
4 of 5 tasks
3
DSV4 mxfp4->nvfp4 cast
#1406 opened May 7, 2026 by cjluo-nv Collaborator • Draft
3
Schematize config loading and quantizer config entries
#1405 opened May 7, 2026 by shengliangxu Collaborator Loading…
11
Working branch for NVFP4 MSE
#1403 opened May 7, 2026 by cjluo-nv Collaborator • Draft
4
[6034518] Fix Conv->Relu->Concat Q/DQ insertion gap
#1398 opened May 6, 2026 by willg-nv Contributor Loading…
13
[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
46
fixes for fused moe (qwen3.6, GLM5.1 + MSE calibration
#1382 opened May 2, 2026 by Fridah-nv Contributor Loading…
37
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…
13
Use trtexec_safe on safety platforms when using remoteAutoTuning
#1378 opened Apr 30, 2026 by dthienan-nv Contributor Loading…
27
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…
40
[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
18
[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
ProTip! Add no:assignee to see everything that’s not assigned.