-
Notifications
You must be signed in to change notification settings - Fork 263
All issues
Issue creation is restricted in this repository
- #3278 · HuiyingLi opened
on Jul 28, 2026 - #2958 · khazic opened
on Jul 7, 2026 7 - #3576 · yuhezhang-ai opened
on Aug 18, 2026
Issues
is:issue state:open
is:issue state:open
Search results
qwen3_omni_moe peft save puts the thinker. prefix outside base_model.model
bugSomething isn't workingSomething isn't workingStatus: Open.#3595 In NVIDIA-NeMo/Automodel;fused expert lora adapters can't load on peft >= 0.19 (ParamWrapper layout flipped)
bugSomething isn't workingSomething isn't workingStatus: Open.#3592 In NVIDIA-NeMo/Automodel;Support dflash2
enhancementNew feature or requestNew feature or requestStatus: Open.#3590 In NVIDIA-NeMo/Automodel;- Status: Open.#3587 In NVIDIA-NeMo/Automodel;
- Status: Open.#3586 In NVIDIA-NeMo/Automodel;
- Status: Open.#3578 In NVIDIA-NeMo/Automodel;
Roadmap: checkpoint I/O architecture, correctness, performance, and memory
enhancementNew feature or requestNew feature or requestStatus: Open.#3576 In NVIDIA-NeMo/Automodel;- Status: Open.#3568 In NVIDIA-NeMo/Automodel;
Investigate TE rope fusion
bugSomething isn't workingSomething isn't workingStatus: Open.#3565 In NVIDIA-NeMo/Automodel;Fix mtp+cp support for existing models
bugSomething isn't workingSomething isn't workingStatus: Open.#3564 In NVIDIA-NeMo/Automodel;- Status: Open.#3520 In NVIDIA-NeMo/Automodel;
- Status: Open.#3519 In NVIDIA-NeMo/Automodel;
