-
Notifications
You must be signed in to change notification settings - Fork 386
- #3754 · anwithk opened
on May 8, 2026
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
[model] qwen3.5
area:modelModel implementations and HF bridge logicModel implementations and HF bridge logicfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workneeds-triageNew item needs classification and ownershipNew item needs classification and ownershipStatus: Open.#4570 In NVIDIA-NeMo/Megatron-Bridge;[bug] Abnormal loss fluctuation when resuming training from checkpoint
bugSomething isn't workingSomething isn't workingneeds-triageNew item needs classification and ownershipNew item needs classification and ownershipStatus: Open.#4565 In NVIDIA-NeMo/Megatron-Bridge;- Status: Open.#4544 In NVIDIA-NeMo/Megatron-Bridge;
[perf] Enable HybridEP for THD layout training
area:trainingTraining loop, callbacks, and runtime integrationTraining loop, callbacks, and runtime integrationfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement worktrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4538 In NVIDIA-NeMo/Megatron-Bridge;PoR: Tinker-like Megatron TrainingEngine APIs
area:trainingTraining loop, callbacks, and runtime integrationTraining loop, callbacks, and runtime integrationfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4515 In NVIDIA-NeMo/Megatron-Bridge;PoR: Megatron FSDP in Qwen3-VL 30B-A3B
area:trainingTraining loop, callbacks, and runtime integrationTraining loop, callbacks, and runtime integrationfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4514 In NVIDIA-NeMo/Megatron-Bridge;PoR: [support] NVFP4 Qautnization-Aware Distillation (QAD) for the following models
area:recipeTraining recipes and launch configsTraining recipes and launch configsfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4512 In NVIDIA-NeMo/Megatron-Bridge;PoR: migrate to MCore HybridModel
area:modelModel implementations and HF bridge logicModel implementations and HF bridge logicfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.POR: Track model support verification levels
area:modelModel implementations and HF bridge logicModel implementations and HF bridge logicfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4509 In NVIDIA-NeMo/Megatron-Bridge;PoR: [model] tracking: GLM-5.2 support
area:modelModel implementations and HF bridge logicModel implementations and HF bridge logicfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4490 In NVIDIA-NeMo/Megatron-Bridge;PoR: Merge MLM dev branch MoE recipes into Bridge training recipes
area:recipeTraining recipes and launch configsTraining recipes and launch configsarea:trainingTraining loop, callbacks, and runtime integrationTraining loop, callbacks, and runtime integrationfeatureNew capabilities, enhancements, or enablement workNew capabilities, enhancements, or enablement workmlm-syncRequires API/behavior sync with upstream Megatron-LM changesRequires API/behavior sync with upstream Megatron-LM changesPoRPlan of record item for roadmap and release trackingPlan of record item for roadmap and release trackingtrackingTracking issue for an ongoing project with smaller stepsTracking issue for an ongoing project with smaller stepsStatus: Open.#4471 In NVIDIA-NeMo/Megatron-Bridge;[support] Qwen3-30B-A3B DGX-B200 MXFP8: 562 TFLOP/s achieved vs documented 619 TFLOP/s - seeking reproduction conditions
area:perfPerformance optimizations and benchmarkingPerformance optimizations and benchmarkingwaiting-on-customerWaiting on the original author to respondWaiting on the original author to respondStatus: Open.#4439 In NVIDIA-NeMo/Megatron-Bridge;