-
Notifications
You must be signed in to change notification settings - Fork 231
All issues
Issue creation is restricted in this repository
- #2952 · akoumpa opened
on Jul 7, 2026 - #2985 · yuhezhang-ai opened
on Jul 9, 2026 - #2958 · khazic opened
on Jul 7, 2026 7
Issues
is:issue state:open
is:issue state:open
Search results
build(deps): upgrade ffpa-attn to 0.2.2 and track CuTeDSL compatibility
enhancementNew feature or requestNew feature or requestStatus: Open.#3224 In NVIDIA-NeMo/Automodel;- Status: Open.#3219 In NVIDIA-NeMo/Automodel;
VLM validation drops the batch dimension for tensor-valued hidden_states
waiting-on-customerWaiting on the original author to respondWaiting on the original author to respondStatus: Open.#3215 In NVIDIA-NeMo/Automodel;fix(vlm): validation crashes with FusedLinearCrossEntropy — out.hidden_states[-1] indexes the batch dim when the model returns a single hidden-state tensor
bugSomething isn't workingSomething isn't workingwaiting-on-customerWaiting on the original author to respondWaiting on the original author to respondStatus: Open.#3206 In NVIDIA-NeMo/Automodel;- Status: Open.#3191 In NVIDIA-NeMo/Automodel;
- Status: Open.#3190 In NVIDIA-NeMo/Automodel;
Support gigatoken
enhancementNew feature or requestNew feature or requestgood first issueGood for newcomersGood for newcomersStatus: Open.feat: add ViSpec vision-language speculative draft training
waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respondStatus: Open.#3172 In NVIDIA-NeMo/Automodel;[VLM] Support pre-extracted video frames (list of image paths) in _preload_media
enhancementNew feature or requestNew feature or requestwaiting-on-customerWaiting on the original author to respondWaiting on the original author to respondStatus: Open.#3168 In NVIDIA-NeMo/Automodel;feat(speculative): validate multimodal speculative decoding training
waiting-on-maintainersWaiting on maintainers to respondWaiting on maintainers to respondStatus: Open.#3167 In NVIDIA-NeMo/Automodel;- Status: Open.#3160 In NVIDIA-NeMo/Automodel;
- Status: Open.#3149 In NVIDIA-NeMo/Automodel;