Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
231 Open 1,329 Closed
231 Open 1,329 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

feat(specdec): warm-start DFlash fine-tuning from exported drafter via dflash_init_checkpoint
#1978 opened Jul 15, 2026 by h-guo18 Contributor • Draft
4
Quality: Insecure subprocess usage in get_system_info.py
#1977 opened Jul 15, 2026 by tomaioo Loading…
4
Add activation-aware local Hessian scale calibration
#1976 opened Jul 14, 2026 by realAsma Contributor • Draft
4
add Qwen3-VL support for DFlash training
#1975 opened Jul 14, 2026 by skierat Contributor Loading…
6
Add MoE support to runtime stats
#1973 opened Jul 14, 2026 by grzegorz-k-karch Contributor • Draft
4
Multi-GPU vLLM benchmarking for runtime stats
#1972 opened Jul 14, 2026 by grzegorz-k-karch Contributor • Draft
4
docs: add announcements landing page
#1971 opened Jul 13, 2026 by ChenhanYu Collaborator • Draft
4
[Examples]: MiniMax-M3 DSpark
#1965 opened Jul 12, 2026 by h-guo18 Contributor Loading…
11
feat(quant): skip max-calib forward when no activation needs data
#1963 opened Jul 10, 2026 by Edwardf0t1 Contributor Loading…
13
Fix per-tensor FP8 weight dequantization
#1962 opened Jul 10, 2026 by chadvoegele Contributor Loading…
5
[Example]: Qwen3-8B online DSpark launcher example (+ DSpark recipe)
#1959 opened Jul 9, 2026 by h-guo18 Contributor Loading…
7
Wan 2.2 T2V A14B diffusers PTQ: two-expert NVFP4-SVDQuant HF checkpoint export
#1957 opened Jul 9, 2026 by jingyu-ml Contributor • Draft
6
docs: add Nemotron 3 Ultra NVFP4 blog to Latest News
#1956 opened Jul 9, 2026 by shljessie Contributor Loading…
4
Support efficient TEGroupedMLP (moe_grouped_gemm=True) for Minitron Pruning
#1955 opened Jul 9, 2026 by kevalmorabia97 Collaborator • Draft
1 task
9
example(eagle3): Qwen3-0.6B EAGLE3 online/offline launcher examples
#1954 opened Jul 9, 2026 by ChenhanYu Collaborator • Draft
3
Support layerwise VLM image calibration in hf_ptq
#1953 opened Jul 9, 2026 by realAsma Contributor • Draft
2
Add fp16 mixed precision softmax
#1950 opened Jul 9, 2026 by kaix-nv Contributor • Draft
4
[OMNIML-5477, OMNIML-5119] Add module-specific AutoQuant search spaces
#1949 opened Jul 9, 2026 by meenchen Contributor Loading…
@meenchen
11
Fix distributed amax error handling
#1944 opened Jul 8, 2026 by realAsma Contributor Loading…
16
[6385267][ONNX][Autocast] Preserve graph-output Cast producers
#1943 opened Jul 8, 2026 by gcunhase Contributor Loading…
5
Add split-k decode kernel for skip softmax
#1941 opened Jul 8, 2026 by kaix-nv Contributor • Draft
4
MBridge VLM distillation / QAD support
#1938 opened Jul 7, 2026 by kevalmorabia97 Collaborator Loading…
23
Fix IndexError during Qwen3.5/3.6 quantization by disabling in_proj_qkv/in_proj_z by default
#1936 opened Jul 7, 2026 by heyparth1 Loading…
1
4
[6403893][ONNX][AutoCast] Fix false-success TRT parse for large ModelProto
#1928 opened Jul 6, 2026 by galagam Contributor Loading…
@galagam
13
test: targeted unit tests for codecov-uncovered lines across eight modules
#1927 opened Jul 6, 2026 by arham766 Contributor Loading…
2
ProTip! Mix and match filters to narrow down what you’re looking for.