Skip to content

Pull requests: NVIDIA/Model-Optimizer

New pull request New
291 Open 1,600 Closed
291 Open 1,600 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

[Fix] Calibrate non-decoder modules during layerwise quantization
#2339 opened Sep 4, 2026 by realAsma Contributor • Draft
13
[Example] Qwen3.5 DSpark · E2E Training Example
#2338 opened Sep 4, 2026 by jinzex Loading…
20
[5565357] Fix SDXL NVFP4 export and performance
#2336 opened Sep 4, 2026 by ajrasane Contributor • Draft
3
Unpack Qwen3.5 MoE routed experts and keep quantized lm_head on Megatron export cherry-pick-0.47.0 Upcoming release
#2334 opened Sep 4, 2026 by kevalmorabia97 Collaborator Loading…
69
[6529944] Fix dynamic NVFP4 TensorRT compatibility
#2333 opened Sep 4, 2026 by ajrasane Contributor • Draft
3
fix(quantization): fix nvfp4 availability check
#2331 opened Sep 4, 2026 by Lee-YNU Loading…
2
Add Parallel Decoding Distillation to FastGen
#2329 opened Sep 4, 2026 by mxinO Contributor • Draft
25
Rename modelopt_recipes/huggingface to model_type with backward-compat alias
#2328 opened Sep 3, 2026 by shengliangxu Collaborator • Draft
4
added new workflow
#2323 opened Sep 3, 2026 by grzegorz-k-karch Contributor Loading…
3
fp32 master weights for the DFlash draft, and keep them across a resume
#2322 opened Sep 3, 2026 by h-guo18 Contributor Loading…
6
[6466805] Add Dynamo ONNX export support for quantized models
#2321 opened Sep 3, 2026 by ajrasane Contributor • Draft
3
Reject unsupported partial-block INT4/W4A8 AWQ export
#2320 opened Sep 2, 2026 by realAsma Contributor Loading…
57
Fix TEGroupedMLP quantizer checkpoint resharding
#2319 opened Sep 2, 2026 by jenchen13 Contributor Loading…
38
[6410139] Fix ONNX AutoCast for large external initializers cherry-pick-0.47.0 Upcoming release
#2317 opened Sep 2, 2026 by ajrasane Contributor Loading…
@ajrasane
14
docs: document speculation profiles and how to produce them
#2316 opened Sep 2, 2026 by yeyu-nvidia Contributor Loading…
14
ar_validate: emit a speculation profile from in-training AR validation
#2315 opened Sep 2, 2026 by yeyu-nvidia Contributor Loading…
13
[6508436] Fix BF16 FP8 ONNX export cherry-pick-0.47.0 Upcoming release
#2314 opened Sep 2, 2026 by ajrasane Contributor Loading…
@ajrasane
30
export: attach speculation_profile.json to exported draft checkpoints
#2313 opened Sep 2, 2026 by yeyu-nvidia Contributor Loading…
15
Add NVFP4 PTQ recipes for zai-org/GLM-5.3-Flash (experts-only and experts + dense MLP)
#2312 opened Sep 2, 2026 by shengliangxu Collaborator Loading…
6
feat(export): support multimodal and MTP models in layerwise export
#2303 opened Sep 1, 2026 by Fridah-nv Contributor Loading…
70
Add the NVFP4 PTQ recipe for Qwen/Qwen3.8-2.4T-A95B
#2302 opened Sep 1, 2026 by shengliangxu Collaborator Loading…
2
Forward kv_cache_free_gpu_memory_fraction to the TensorRT-LLM engines (NVBug 6701763)
#2300 opened Sep 1, 2026 by cjluo-nv Collaborator Loading…
6
Fix fsdp2_aware_weight_update masking setup errors with UnboundLocalError
#2295 opened Sep 1, 2026 by harshal-96 Loading…
2 of 4 tasks
1
5
feat(quantization): scoped calibration pipelines via algo_cfg [prototype]
#2292 opened Sep 1, 2026 by Fridah-nv Contributor • Draft
4
specdec: config_overrides for nested text_config checkpoints + load VLM-capable bases in merge_lora
#2289 opened Aug 31, 2026 by yeyu-nvidia Contributor Loading…
43
ProTip! Exclude everything labeled bug with -label:bug.