Repository Issue Activity (beta)

#2055 [Bug] NCCL (NVIDIA Collective Communications Library) timeout error when train Qwen3-Coder-30B-A3B-Instruct

#2571 verl tool adapter: synchronous env execution blocks the agent-loop event loop

#1457 [Feature] Add support for token-based rate-limiting in RemoteInferenceEngine

#2106 [Feature] Include Support for Trainium and Inferentia

#2096 [Feature] Support passing tool-use parameters to OpenAI-compatible client in Oumi inference config

#1322 [Feature] Unsloth fine-tuning support

#2053 [Bug] Turning down stopped cluster doesn't work

#1447 [OP][Feature] Add support for loading pre-trained custom models

#1519 [Feature] Add loading spinner during interactive inference

#2067 [Feature] Add k8s support for oumi launcher

#1677 [Bug] Evaluation accuracy drops when using distributed accelerate launch with multi-GPU

#1337 [Feature] Kubernetes Support

#2015 [Bug] Oumi wrapper fail to parse SLURM nodelist when submitting multiple nodes jobs

#2007 [Bug] Can't run distributed training on multiple nodes with slurm using the oumi distributed wrapper.

#1540 [Bug] `training.include_performance_metrics` does not work for GRPO

#1735 [Feature] View output of `oumi launch up` on Slurm job

#1338 [Feature] Qwen2.5-VL SFT + DPO + PPO with Sequence Parallelism

#1980 [Bug] Emptying inference result cache?

#1973 可视化界面

#1629 [Feature] Print out OmegaConf configs at the start of evaluation/inference

#1667 [Feature] Support Llama Maverick tuning and evaluation

#1628 [Feature] Gracefully handle saving OmegaConf configs containing non-primitives

#1658 [Feature] Exploring support for TextGrad and DSPy libraries

#1971 [Feature] Bedrock ConverseAPI support for inference

#1984 Missing Hyper-Parameters tuning module/support