Skip to content

Pull requests: NVIDIA-NeMo/Automodel

New pull request New
153 Open 2,299 Closed
153 Open 2,299 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: deterministic implementation of _float32_rms_norm_fwd
#2971 opened Jul 8, 2026 by ashors1 Loading…
3 tasks
5
docs(fern): cut v0.5.0 version train
#2970 opened Jul 8, 2026 by lbliii Contributor Loading…
4
chore: CI proxy for #2918
#2967 opened Jul 8, 2026 by HuiyingLi Contributor Loading…
2
feat(speculative): add multi-dataset acceptance-length benchmark sweep community-request
#2966 opened Jul 8, 2026 by khazic Contributor Loading…
1
chore: CI proxy for #2953
#2965 opened Jul 8, 2026 by HuiyingLi Contributor Loading…
3
test(transformers): add source-load parity coverage
#2960 opened Jul 7, 2026 by yuhezhang-ai Contributor • Draft
3 tasks done
2
feat(dspark): log acceptance and confidence metrics during training community-request
#2957 opened Jul 7, 2026 by kashif Contributor Loading…
4
fix(moe): preserve MLP dispatch through checkpoint wrappers
#2955 opened Jul 7, 2026 by akoumpa Contributor • Draft
3 tasks done
6
feat(kd): support separate student and teacher meshes
#2954 opened Jul 7, 2026 by akoumpa Contributor • Draft
3 tasks done
3
perf(moe): page partial graph expert activations
#2944 opened Jul 6, 2026 by yaoyu-33 • 4/4 • Draft
1
perf(cuda-graphs): share partial capture pools
#2943 opened Jul 6, 2026 by yaoyu-33 • 3/4 • Draft
1
fix(transformers): gate _tie_weights_nemo on tie_word_embeddings flag
#2942 opened Jul 6, 2026 by yuhezhang-ai Contributor Loading…
4 tasks done
1
2
ci: AUT-707 bump FW-CI-templates _claude_review.yml to v1.7.1 ci
#2938 opened Jul 6, 2026 by svcnemo-autobot Loading…
6
refactor(distributed): unify CP input prep and dispatch across models
#2937 opened Jul 6, 2026 by HuiyingLi Contributor • Draft
2 of 3 tasks
1
2
feat(cuda-graphs): capture fixed-capacity MoE execution
#2932 opened Jul 6, 2026 by yaoyu-33 • 2/4 • Draft
2 of 3 tasks
1
fix(cp): preserve gradients when sharding VLM inputs
#2931 opened Jul 6, 2026 by HuiyingLi Contributor Loading…
12 tasks done
7
feat(moe): DeepEP v2 dispatcher, Torch 2.12 PP, and selectable EP variants
#2930 opened Jul 5, 2026 by hemildesai Contributor Loading…
3 tasks done
3
feat(attention): support flash_attention_3 and flash_attention_4
#2929 opened Jul 5, 2026 by HuiyingLi Contributor Loading…
3
feat(pp): support FusedLinearCrossEntropy under pipeline parallelism community-request
#2927 opened Jul 5, 2026 by beccohov Loading…
2 of 3 tasks
1
4
ci: Update transformers to latest version 5.13.0
#2925 opened Jul 5, 2026 by svcnvidia-nemo-ci Contributor Loading…
1
feat(speculative): context parallelism for draft-model training community-request
#2918 opened Jul 3, 2026 by kashif Contributor • Draft
32
perf(moe): add scoped partial CUDA graphs
#2917 opened Jul 3, 2026 by yaoyu-33 • 1/4 • Draft
2
feat(retrieval): add mining data preparation utilities
#2907 opened Jul 3, 2026 by oliverholworthy Contributor • Draft
4 tasks done
1
fix(retrieval): preserve bi-encoder export metadata
#2905 opened Jul 3, 2026 by oliverholworthy Contributor • Draft
4 tasks done
1
fix(retrieval): harden hard-negative mining
#2906 opened Jul 3, 2026 by oliverholworthy Contributor • Draft
4 tasks done
1
ProTip! Mix and match filters to narrow down what you’re looking for.