-
Notifications
You must be signed in to change notification settings - Fork 273
Pull requests: NVIDIA-NeMo/Automodel
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
feat(recipes): add model-ready hooks and checkpoint addons
#3661
opened Aug 25, 2026 by
pstjohn
Loading…
Enable activation checkpointing for repeated dense MTP blocks
#3660
opened Aug 25, 2026 by
pzelasko
Contributor
Loading…
fix(test): unblock the Kimi-Linear vanilla-HF parity reference
#3659
opened Aug 25, 2026 by
yuhezhang-ai
Contributor
Loading…
perf(dllm): use the DeepEP dispatcher in the DiffusionGemma ep=8 recipes
#3654
opened Aug 25, 2026 by
akoumpa
Contributor
Loading…
feat(registry): add public architecture registration and entry-point discovery
#3645
opened Aug 24, 2026 by
pstjohn
Loading…
fix(minimax): repair HF parity reference and gate at the measured envelope
#3643
opened Aug 24, 2026 by
yuhezhang-ai
Contributor
•
1/2
Loading…
fix(moe): equalize and align per-rank token counts for HybridEP dispatch
#3641
opened Aug 24, 2026 by
HuiyingLi
Contributor
Loading…
feat(laguna): support packed THD context parallelism
#3640
opened Aug 24, 2026 by
akoumpa
Contributor
Loading…
3 tasks done
fix: Canonicalize LoRA compute dtype
community-request
#3638
opened Aug 24, 2026 by
benthecarman
Loading…
3 tasks done
fix(glm): align and diagnose cross-framework router parity
#3635
opened Aug 22, 2026 by
yuhezhang-ai
Contributor
•
3/3
•
Draft
chore: bump lint tools + fix lint issues
community-request
#3634
opened Aug 22, 2026 by
akx
Contributor
Loading…
3 tasks done
fix(kimi_k25): honour the quantization flag
community-request
waiting-on-customer
Waiting on the original author to respond
#3633
opened Aug 22, 2026 by
akx
Contributor
Loading…
3 tasks done
fix(qwen3_omni_moe): put the thinker prefix inside the peft prefix on lora saves
community-request
#3630
opened Aug 22, 2026 by
stanley1208
Contributor
Loading…
ci: Update transformers to latest version 5.15.1
#3628
opened Aug 22, 2026 by
svcnvidia-nemo-ci
Contributor
Loading…
feat(dflash): support top-p and top-k in speculative decoding
community-request
waiting-on-customer
Waiting on the original author to respond
#3627
opened Aug 22, 2026 by
kashif
Contributor
Loading…
3 tasks done
fix(moe): export fused expert lora in the corrected peft >= 0.19.1 layout
community-request
waiting-on-customer
Waiting on the original author to respond
#3626
opened Aug 22, 2026 by
stanley1208
Contributor
Loading…
fix(utils): accumulate param L2 norm in float32 for bf16/fp16 models
community-request
#3625
opened Aug 22, 2026 by
ralovets
Contributor
Loading…
2 of 3 tasks
perf(checkpoint): bound GPT-OSS MXFP4 loading
#3623
opened Aug 22, 2026 by
yuhezhang-ai
Contributor
•
5/5
•
Draft
perf(checkpoint): avoid full GC scans during MoE export
#3621
opened Aug 22, 2026 by
yuhezhang-ai
Contributor
Loading…
test(ci): report Jensen-Shannon checkpoint parity metrics
#3620
opened Aug 22, 2026 by
yuhezhang-ai
Contributor
•
2/3
Loading…
3 tasks done
perf(checkpoint): bound quantized DCP load memory
#3619
opened Aug 21, 2026 by
yuhezhang-ai
Contributor
•
4/5
•
Draft
Previous Next
ProTip!
no:milestone will show everything without a milestone.