-
Notifications
You must be signed in to change notification settings - Fork 360
Pull requests: NVIDIA/Model-Optimizer
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[minor] Add custom calibration backend registry
#1281
opened Apr 16, 2026 by
Fridah-nv
Contributor
Loading…
Fix and Speedup megatron_mmlu by >10x via prefill scoring and global batching
#1280
opened Apr 16, 2026 by
kevalmorabia97
Collaborator
Loading…
feat(serialization): remove explicit weights_only default from safe_load
#1279
opened Apr 16, 2026 by
kevalmorabia97
Collaborator
Loading…
1 task done
Update 0.43 release date in CHANGELOG.rst
#1277
opened Apr 16, 2026 by
kevalmorabia97
Collaborator
Loading…
Removed version fixes for torch transformers in windows ptq example requirements
#1275
opened Apr 16, 2026 by
hthadicherla
Contributor
Loading…
[Feat,Refactor]: Offline Dflash; Spec Mixin; Deprecate parallel draft;
#1271
opened Apr 16, 2026 by
h-guo18
Contributor
Loading…
Centralize 'trtexec' subprocess runs in ONNX into a single function
#1268
opened Apr 15, 2026 by
gcunhase
Contributor
Loading…
Handle zero-amax per-channel activation scaling for MoE export
#1265
opened Apr 15, 2026 by
AEON-7
Loading…
Fix non-scalar input amax in preprocess_linear_fusion for MoE export
#1264
opened Apr 15, 2026 by
AEON-7
Loading…
Add ResNet50 support for torch_onnx quantization workflow
#1263
opened Apr 14, 2026 by
ajrasane
Contributor
Loading…
2 tasks done
Exclude small-k and small-n Matmul nodes from Int8 quantization
#1256
opened Apr 14, 2026 by
nv-samcheng
Contributor
Loading…
Add EfficientViT support for torch_onnx quantization workflow
#1254
opened Apr 14, 2026 by
ajrasane
Contributor
Loading…
3 tasks done
Add a general composable $import system for YAML configs, and use it to implement composable recipes
#1253
opened Apr 14, 2026 by
shengliangxu
Collaborator
Loading…
Add a standalone monitor skill for persistent job tracking
#1252
opened Apr 14, 2026 by
kaix-nv
Contributor
Loading…
Add layerwise calibration for large models
#1251
opened Apr 13, 2026 by
realAsma
Contributor
Loading…
2 of 4 tasks
fix(launcher): use afterany dependency for allow_to_fail pipelines
#1248
opened Apr 13, 2026 by
yeyu-nvidia
Contributor
Loading…
3 tasks
Add LAQ (Learnable Amax Quantization) algorithm
#1247
opened Apr 13, 2026 by
realAsma
Contributor
Loading…
4 tasks
vLLM fakequant export update for AWQ checkpoint
#1242
opened Apr 13, 2026 by
kinjalpatel27
Contributor
Loading…
feat: parallelize fakequant export across GPUs via ThreadPoolExecutor
#1241
opened Apr 13, 2026 by
kinjalpatel27
Contributor
Loading…
[1/N] Polish evaluation skills and common skills based on an E2E workflow testing
#1239
opened Apr 12, 2026 by
Edwardf0t1
Contributor
Loading…
Previous Next
ProTip!
no:milestone will show everything without a milestone.