Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
256 Open 1,497 Closed
256 Open 1,497 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(speculative): support Gemma-4-E4B as a streaming DFlash/DSpark target
#2186 opened Aug 13, 2026 by h-guo18 Contributor Loading…
8
fix: guard MoE branch when quantization_format is None
#2185 opened Aug 13, 2026 by andrewwhitecdw Loading…
2
Add Aumann-Shapley sensitivity scoring method to auto_quantize
#2183 opened Aug 12, 2026 by joshua-hill • Draft
2
Remove deprecations marked in 0.45 and 0.46
#2182 opened Aug 12, 2026 by kevalmorabia97 Collaborator Loading…
40
[OMNIML-5563] Add PETR ONNX PTQ and accuracy evaluation example
#2180 opened Aug 12, 2026 by ajrasane Contributor Loading…
24
Export quantized/co-trained MTP weights instead of copying BF16
#2174 opened Aug 12, 2026 by yeyu-nvidia Contributor Loading…
11
Express the Cosmos3 Nano DFlash recipe as a launcher YAML example
#2173 opened Aug 12, 2026 by h-guo18 Contributor Loading…
2
Automate Puzzletron GPU lifecycle tests
#2167 opened Aug 12, 2026 by j-rausch Contributor Loading…
34
Add Puzzletron v2 GPU quality baseline
#2166 opened Aug 12, 2026 by j-rausch Contributor Loading…
69
examples: add Qwen3-8B streaming DSpark launcher example + walkthrough
#2164 opened Aug 12, 2026 by h-guo18 Contributor Loading…
5
Fix init_quantized_weights: model kwargs leak into dispatch, missing tie_weights()
#2161 opened Aug 12, 2026 by spped2000 Loading…
14
Simplify tied-weight export dedup
#2151 opened Aug 11, 2026 by chadvoegele Contributor • Draft
4
Support fine-tuning released DFlash/DSpark drafters (causal SWA, attention sink, warm start)
#2149 opened Aug 11, 2026 by h-guo18 Contributor • Draft
4
Add reversible weight folding and support TE grouped weights
#2140 opened Aug 11, 2026 by mxinO Contributor Loading…
31
HAQ Poc
#2138 opened Aug 11, 2026 by juhi10071998 Contributor • Draft
@juhi10071998
14
feat(export): export each decoder layer as layerwise calibration finishes it
#2136 opened Aug 10, 2026 by Fridah-nv Contributor • Draft
5
Add Inkling attention support to vLLM runtime
#2135 opened Aug 10, 2026 by meenchen Contributor • Draft
@meenchen
2
Skip weight quantizers with no stored weight in fold_weight
#2132 opened Aug 10, 2026 by babyplutokurt Loading…
1
12
PTQ reliability fixes: checkpoint resume + exclude_modules sentinel corruption
#2129 opened Aug 10, 2026 by wyattearp Loading…
9
specdec_bench: fix run provenance, add MLflow tracking
#2119 opened Aug 10, 2026 by h-guo18 Contributor Loading…
44
Fix AA-next OpenHands timeout guidance
#2094 opened Aug 6, 2026 by Edwardf0t1 Contributor Loading…
8
Add id/ tied_weight_names based export shared weight deduplication logic and remove legacy data_ptr approach
#2092 opened Aug 6, 2026 by juhi10071998 Contributor Loading…
78
Fix clip backward with broadcasted constant bounds
#2089 opened Aug 6, 2026 by Robertboy18 Loading…
3
refactor(export): split unified_export_hf into layered modules
#2088 opened Aug 6, 2026 by Fridah-nv Contributor Loading…
32
Add FastGen quantization-aware distillation
#2085 opened Aug 5, 2026 by jingyu-ml Contributor • Draft
2
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.