microsoft/agent-lightning

How to evaluate the trained model? 18 days ago

help wanted question verl v0.x

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

AgentOps incorrect message capture causes KeyError in convert_to_openai_messages 29 days ago

help wanted need investigation adapter v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Cleaner termination 29 days ago

help wanted store v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

How to deal with failed rollouts? 29 days ago

documentation help wanted question v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Which hyper-parameters in the standard VERL configuration do not take effect? 29 days ago

documentation enhancement help wanted v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

[Over-long prompt/response length] cannot reshape tensor of 0 elements into shape [1, 0, -1, 128] because the unspecified dimension size -1 can be any value and is ambiguous 29 days ago

help wanted verl v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Is is possible to support Google Agent Development Kit (ADK) in the future? 29 days ago

help wanted examples v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

ValueError: `position_ids should be a 3D tensor of shape (4, batch_size, seq_length)` when using agent-lightning 0.2.1 + Qwen2.5-VL 29 days ago

help wanted topic/multimodality v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Error with vllm 29 days ago

help wanted verl out-of-memory vllm v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Completed rollouts > total_tasks_queued 29 days ago

help wanted verl v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

How to assign different LoRA to different sub-agents? 29 days ago

help wanted verl v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

RFC: Integrate Slime RL backend into Agent Lightning (tinker-style service vs native integration) 29 days ago

enhancement help wanted v0.x

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

How can agent built with Java (Spring AI) integrate with agent lightning? 8 months ago

help wanted

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Support for TS/JS based agent frameworks? 8 months ago

help wanted

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

[Feature] Add data validation steps to `trainer.dev` for early error detection 10 months ago

enhancement help wanted verl

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

BUG: Compatibility with verl 0.6.0 and LoRA 10 months ago

bug help wanted verl

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Sequence ID in dashboard won't sort properly 10 months ago

help wanted dashboard

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

langchain 1.0 needed update 10 months ago

help wanted

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Hitting KeyboardInterrupt produces too many error logs 10 months ago

help wanted

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

I find agent lightning only use 1 actor during rollout, can we launch multiple actors? 10 months ago

help wanted question verl

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

How to run agentic rl with agent lightning in the distributed training environment with multi-nodes and multi-gpus ? 10 months ago

help wanted question examples verl

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Is is possible to run training task on an agent built by Golang? 10 months ago

help wanted

microsoft/agent-lightning

17.7K

The absolute trainer to light up AI agents.

Python

#agent#agentic-ai#llm#mlops#reinforcement-learning

Open Issues Need Help