Search results

Support for MLLM

Status: Open.

#9 In THUDM/AgentRL;

· jangyicheng opened on Jan 7, 2026

Request for AgentRL Multi-Task / Single-Task Model Weights

Status: Open.

#8 In THUDM/AgentRL;

· ArkaZhuo opened on Dec 18, 2025

NCCL Error: "Cuda failure 'invalid argument'" in NCCLTensorReceiver.async_receive()

Status: Open.

#7 In THUDM/AgentRL;

· Gikiman opened on Nov 16, 2025

模型问题

Status: Open.

#6 In THUDM/AgentRL;

· ZY-LUCK opened on Nov 8, 2025

Approx resource requirements

Status: Open.

#5 In THUDM/AgentRL;

· 0xrushi opened on Nov 8, 2025

Difference with the Slime project

Status: Open.

#4 In THUDM/AgentRL;

· dangerzone opened on Nov 1, 2025

The Question about Cross-Policy Sampling Strategy

Status: Open.

#3 In THUDM/AgentRL;

· supermancmk opened on Oct 23, 2025