-
Notifications
You must be signed in to change notification settings - Fork 196
Pull requests: NVIDIA-NeMo/Automodel
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(diffusion_gemma): make parity test work on transformers >= 5.11
#2883
opened Jul 1, 2026 by
zyzhou5
Contributor
Loading…
feat(retrieval): add optimized Nemotron VL custom path
#2872
opened Jun 30, 2026 by
yuhezhang-ai
Contributor
•
Draft
ci(automodel): set AM-576 release timeouts
r0.5.0
Auto-cherrypick to release branch. Apply before merge; cherrypick happens after merge.
#2871
opened Jun 30, 2026 by
yuhezhang-ai
Contributor
Loading…
feat: support AutoModelForSeq2SeqLM (encoder-decoder fine-tuning)
community-request
waiting-on-customer
Waiting on the original author to respond
#2860
opened Jun 30, 2026 by
stanley1208
Contributor
Loading…
build(docker): add selectable EP CUDA image
#2855
opened Jun 29, 2026 by
hemildesai
Contributor
•
6/6
•
Draft
perf(moe): reuse HybridEP JIT cache
#2854
opened Jun 29, 2026 by
hemildesai
Contributor
•
5/6
•
Draft
test(moe): add DeepEP v2 benchmark and finetuning recipes
#2853
opened Jun 29, 2026 by
hemildesai
Contributor
•
4/6
•
Draft
fix(pp): adapt pipeline execution to Torch 2.12
#2851
opened Jun 29, 2026 by
hemildesai
Contributor
•
3/6
•
Draft
build(moe): add selectable DeepEP dependency variants
#2850
opened Jun 29, 2026 by
hemildesai
Contributor
•
1/6
•
Draft
fix(distributed): fix and scope vlm activation checkpointing
#2840
opened Jun 29, 2026 by
yuhezhang-ai
Contributor
Loading…
fix(loss): localize DTensor labels before KD masking
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#2815
opened Jun 27, 2026 by
fallintoplace
Loading…
fix(config): replace assert-based runtime validation
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#2814
opened Jun 27, 2026 by
fallintoplace
Loading…
3 tasks done
fix(config): enforce safe file target boundaries
community-request
waiting-on-maintainers
Waiting on maintainers to respond
#2813
opened Jun 27, 2026 by
fallintoplace
Loading…
3 tasks done
feat(nemotron_v3): run MTP heads under eval for validation acceptance metrics
#2802
opened Jun 26, 2026 by
Slyne
Contributor
Loading…
2 of 3 tasks
ci(dllm): add dLLM SFT nightly train-to-generate launcher and recipes
#2783
opened Jun 25, 2026 by
zyzhou5
Contributor
Loading…
feat(moe): add DeepEP v2 dispatcher support
#2782
opened Jun 25, 2026 by
hemildesai
Contributor
•
2/6
•
Draft
fix(distributed): control frozen multimodal FSDP sharding
#2763
opened Jun 25, 2026 by
yuhezhang-ai
Contributor
•
Draft
fix(qwen3-moe): export grouped HF expert weights
r0.5.0
Auto-cherrypick to release branch. Apply before merge; cherrypick happens after merge.
ci: update claude review guidelines
#2739
opened Jun 23, 2026 by
akoumpa
Contributor
Loading…
3 tasks
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.