Search results

Remove silent judge error when custom prompt less than 50 characters

Status: Open.

#59 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on Nov 7, 2025

Upgrade litellm to not include [proxy] package; allow upgraded uvicorn

Status: Open.

#53 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on Oct 21, 2025

Vllm server for reward model is broken in 'vllm=0.10.0' transformers=0.53.2

Status: Open.

#43 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on Oct 1, 2025

Reward_Hub should double check to ensure no duplicate system prompt is added / or is properly handled.

Status: Open.

#39 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on Sep 18, 2025

Understand why reward hub don’t support latest vllm version

Status: Open.

#35 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on Sep 3, 2025

verifier interface design

Status: Open.

#32 In Red-Hat-AI-Innovation-Team/reward_hub;

· xukai92 opened on Aug 11, 2025

support math eval OOB

Status: Open.

#31 In Red-Hat-AI-Innovation-Team/reward_hub;

· xukai92 opened on Jun 18, 2025

Add error handling for openai api calls; eg. long context length crashes server

Status: Open.

#23 In Red-Hat-AI-Innovation-Team/reward_hub;

· gx-ai-architect opened on May 1, 2025

make sure the score from different RMs are consistent for before/after sigmoid

Status: Open.

#4 In Red-Hat-AI-Innovation-Team/reward_hub;

· xukai92 opened on Apr 23, 2025