Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
68 Open 451 Closed
68 Open 451 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

Fixes for Megatron Expert Parallel, GroupedMLP and SequentialMLP
#831 opened Jan 30, 2026 by realAsma Loading…
8
Latent MOE & Repeated MTP support for NemotronH; fix KV cache quant export
#830 opened Jan 29, 2026 by jenchen13 Loading…
14
Noeyy/add test cases for the newly added checkpoints on HF
#827 opened Jan 29, 2026 by noeyy-mino Loading…
@noeyy-mino
6
hardcode support for qwen3vl text only
#826 opened Jan 28, 2026 by h-guo18 • Draft
3
[Minor] fix: do not requantize the scales in FP8 scale sweep calibration
#825 opened Jan 28, 2026 by Fridah-nv Loading…
@Fridah-nv
4
Update on the QuantModule & DynamicModule to accept external forward
#824 opened Jan 28, 2026 by jingyu-ml Loading…
@jingyu-ml
5
[ONNX][Autocast] Minor bug fixes (AI-assisted)
#822 opened Jan 28, 2026 by galagam Loading…
2
Support Kimi-K2.5 PTQ
#820 opened Jan 27, 2026 by Edwardf0t1 • Draft
8
Support MiniMax M2.1 (FP8 checkpoint)
#817 opened Jan 25, 2026 by cjluo-nv Loading…
6
Fix mcore nvfp4 export for vllm
#816 opened Jan 24, 2026 by meenchen Loading…
@meenchen
2
Drop '_pg_collection' in MBridge model config when checkpointing
#813 opened Jan 23, 2026 by AAnoosheh • Draft
@AAnoosheh
3
Added column-major storage of weights and scales in INT4 quantization for model load time improvement in TRT-RTX
#811 opened Jan 23, 2026 by hthadicherla Loading…
16
[2/4] Diffusion Quantized ckpt export
#810 opened Jan 23, 2026 by jingyu-ml Loading…
2 of 4 tasks
@jingyu-ml
25
add value info to the original tensor if it was directly to model output
#807 opened Jan 22, 2026 by YixuanSeanZhou • Draft
2
vllm fakequant reload with modelopt state for HF
#805 opened Jan 21, 2026 by kinjalpatel27 Loading…
12
Layerwise KD mode
#802 opened Jan 21, 2026 by AAnoosheh Loading…
@AAnoosheh
15
Add Megatron-Bridge pruning example scripts
#800 opened Jan 21, 2026 by kevalmorabia97 Loading…
1 of 2 tasks
16
[Do not merge] test int4 dequant kernel
#798 opened Jan 20, 2026 by cjluo-nv • Draft
1
GLM-4.7 MTP support
#792 opened Jan 16, 2026 by Edwardf0t1 • Draft
2
add local hessian calibration
#788 opened Jan 16, 2026 by Fridah-nv • Draft
@Fridah-nv
5
Add Nemotron parse PTQ support
#786 opened Jan 15, 2026 by Edwardf0t1 Loading…
2
Finally FLUX NVFP4 quantization working
#782 opened Jan 14, 2026 by FurkanGozukara Loading…
8
[For RL] Keep attrs after folding weight and fix empty extra state for Megatron
#779 opened Jan 14, 2026 by mxinO • Draft
3
Support multiple-batch input for autocast calibration.
#760 opened Jan 11, 2026 by byte-deve Loading…
16
[draft] bug for MoE distributed parallelism
#752 opened Jan 8, 2026 by realAsma • Draft
10
ProTip! Exclude everything labeled bug with -label:bug.