Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
253 Open 1,411 Closed
253 Open 1,411 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

docs: rebuild the unified HF deployment support matrix from the deploy test suite (NVBug 6550792)
#2087 opened Aug 5, 2026 by Edwardf0t1 Contributor Loading…
5
Consolidate example CI lanes and bump containers to latest
#2086 opened Aug 5, 2026 by kevalmorabia97 Collaborator Loading…
4
Add FastGen quantization-aware distillation
#2085 opened Aug 5, 2026 by jingyu-ml Contributor • Draft
2
[NVBUG: 6562021] Fix vLLM FlashAttention KV cache layout handling
#2084 opened Aug 5, 2026 by sychen52 Contributor Loading…
@sychen52
17
Add FP8 Vision Encoder quantization for Qwen3-VL and Qwen3.5
#2083 opened Aug 5, 2026 by mpariente-nvda Loading…
6
[6562078]: fix calibration for vLLM 0.26.0 cherry-pick-0.46.0
#2082 opened Aug 5, 2026 by kinjalpatel27 Contributor Loading…
33
Fix/tied weight export identity
#2081 opened Aug 5, 2026 by chadvoegele Contributor Loading…
9
fix(specdec): correct resume and bound staging in the vLLM hidden-state dump
#2080 opened Aug 5, 2026 by yeyu-nvidia Contributor Loading…
14
fix(autotune): pre-check remote board connectivity before benchmark
#2078 opened Aug 5, 2026 by willg-nv Contributor Loading…
10
Puzletron v2 dockerfile
#2077 opened Aug 5, 2026 by chochowski Contributor Loading…
6
fix(export): honor sub-model scope_prefix in quant-aware reverse rename (NVBug 6525511) cherry-pick-0.46.0
#2076 opened Aug 5, 2026 by yueshen2016 Contributor Loading…
19
feat(megatron-bridge): adopt TEGroupedMLP per-expert quantizers (#1550) for W4A16 NVFP4 four_over_six
#2072 opened Aug 5, 2026 by yueshen2016 Contributor Loading…
17
fix(export): save models with legacy list-style _tied_weights_keys (NVBug 6518665) cherry-pick-0.46.0
#2071 opened Aug 5, 2026 by Edwardf0t1 Contributor Loading…
9
Add Qwen-Image DMD2 QAT and PEFT-backed SVDQuant
#2069 opened Aug 5, 2026 by jingyu-ml Contributor • Draft
4
Use lm-eval 0.4.12's built-in trtllm backend, deprecate lm_eval_tensorrt_llm.py
#2066 opened Aug 4, 2026 by cjluo-nv Collaborator Loading…
39
Bug fix: 6542481 cherry-pick-0.46.0
#2064 opened Aug 4, 2026 by sugunav14 Contributor Loading…
39
[skill] evaluation: align nel-next TB2.1/SWE-bench with golden toolchain
#2063 opened Aug 4, 2026 by cjluo-nv Collaborator • Draft
3
Preserve HF PTQ checkpoint sidecar files [NV BUG 6491822] cherry-pick-0.46.0
#2060 opened Aug 4, 2026 by jenchen13 Contributor Loading…
46
Fix FSDP2 handling for tied embeddings
#2059 opened Aug 3, 2026 by realAsma Contributor • Draft
8
docs(diffusers): clarify --calib-size is samples/prompts, not steps
#2058 opened Aug 3, 2026 by felipeofdev-ai Loading…
1 task
1
3
feat(megatron-bridge): Nemotron-Nano-3 W4A16 NVFP4 four_over_six PTQ/QAD support
#2055 opened Aug 3, 2026 by yueshen2016 Contributor Loading…
23
Add Cosmos3 Nano DFlash multimodal training recipe
#2053 opened Aug 3, 2026 by skierat Contributor Loading…
24
Kimi-K3 support for speculative-decoding draft training
#2049 opened Aug 3, 2026 by h-guo18 Contributor • Draft
3
[chore]: weekly bump of uv.lock on main (2026-08-03)
#2048 opened Aug 3, 2026 by github-actions Bot Loading…
2
Add vLLM skip-softmax and mask-reuse calibration
#2045 opened Aug 1, 2026 by kaix-nv Contributor • Draft
4
ProTip! Type g i on any issue or pull request to go back to the issue listing page.