Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
110 Open 618 Closed
110 Open 618 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

Step3.5 MoE support
#1063 opened Mar 17, 2026 by meenchen • Draft
6
Remove _moe_count_expert_calib_tokens flag; tie token counting to moe_calib_experts_ratio
#1062 opened Mar 17, 2026 by cjluo-nv Loading…
1 task done
13
[minor] Refactor TE fused-norm handling in GPTModelExporter
#1061 opened Mar 17, 2026 by yueshen2016 Loading…
2
Add LoRA co-training support for HF EAGLE speculative decoding
#1060 opened Mar 17, 2026 by yeyu-nvidia Loading…
7
Protobuf 7.34 fix for model size > 2gib
#1058 opened Mar 17, 2026 by hthadicherla Loading…
5
Fused QKV add node issue for GQA graph surgery
#1057 opened Mar 17, 2026 by hthadicherla Loading…
2
security(opt): enable weights_only=True by default
#1056 opened Mar 17, 2026 by RinZ27 Loading…
2
security(peft): enforce torch.nn.init prefix validation
#1054 opened Mar 17, 2026 by RinZ27 Loading…
2
Add VSA
#1053 opened Mar 17, 2026 by kaix-nv Loading…
5
[Feat]FakeBaseModel for offline eagle; Kimi-K2.5 fixes;
#1052 opened Mar 17, 2026 by h-guo18 • Draft
@h-guo18
3
[Do Not merge]Draft:Add NVFP4 four-over-six (4o6) adaptive activation quantization
#1050 opened Mar 16, 2026 by Fridah-nv • Draft
@Fridah-nv
3
MR branch for the remaining difference between dkorzekwa/any_model an…
#1047 opened Mar 16, 2026 by danielkorzekwa Loading…
1
[EAGLE] Optimize EAGLE Training
#1044 opened Mar 16, 2026 by benchislett Loading…
25
Add calib_include/exclude_modules to calibration algorithms
#1043 opened Mar 16, 2026 by Fridah-nv Loading…
@Fridah-nv
18
[EAGLE] Configurable number of TTT steps
#1042 opened Mar 16, 2026 by benchislett Loading…
5
Add MoE TEGroupedMLP support for Minitron Pruning
#1038 opened Mar 13, 2026 by kevalmorabia97 • Draft
1 of 3 tasks
3
Fix FP8 ONNX export to use standard QuantizeLinear/DequantizeLinear ops
#1037 opened Mar 13, 2026 by IgorBaratta Loading…
10
Merge mbridge distillation for any_model
#1036 opened Mar 13, 2026 by danielkorzekwa Loading…
13
Merge any_model tutorial
#1035 opened Mar 13, 2026 by danielkorzekwa Loading…
31
add: ModelOpt Launcher for Slurm job submission
#1031 opened Mar 12, 2026 by ChenhanYu Loading…
4 of 5 tasks
@ChenhanYu
71
Add experts_only mode
#1030 opened Mar 12, 2026 by cjluo-nv • Draft
2
Torch PTQ ONNX export example add for windows + small fixes to existing torch onnx export path
#1027 opened Mar 11, 2026 by hthadicherla Loading…
14
Add Full TE Spec support for Megatron Pruning DynamicModules + MoE bug fixes
#1024 opened Mar 11, 2026 by kevalmorabia97 Loading…
4 of 5 tasks
5
Activation hooks redesign (reuse hooks component across both minitron and puzzletron)
#1022 opened Mar 11, 2026 by danielkorzekwa Loading…
27
Added 2:4 sparsity to skip softmax method
#1019 opened Mar 11, 2026 by rohansjoshi Loading…
7
ProTip! Mix and match filters to narrow down what you’re looking for.