List your company

Multi-turn task completion on SearchGym

92.8Task Completion ScoreGemini-1.5-Pro
32.89648.4486479.552Apr 8, 2026

Evaluation results

MethodLinks
Gemini-1.5-ProCategory=Closed-Source LLM2026.0492.8
Gemini-1.5-FlashCategory=Closed-Source LLM2026.0492.8
Qwen2.5-14B + PRIMECategory=PRIME Models2026.0489.2
GRPO + Qwen2.5-8BCategory=RL (GRPO) Trained Models2026.0488.8
GPT-4oCategory=Closed-Source LLM2026.0488
Qwen2.5-8B + PRIMECategory=PRIME Models2026.0487.2
GRPO + Qwen2.5-4BCategory=RL (GRPO) Trained Models2026.0486.4
Qwen2.5-4B + PRIMECategory=PRIME Models2026.0486
Qwen2.5-4BCategory=Open-Source LLM2026.0485.6
Qwen2.5-8BCategory=Open-Source LLM2026.0480.8
Qwen2.5-32BCategory=Open-Source LLM2026.0479.2
Qwen2.5-14BCategory=Open-Source LLM2026.0451.2
GPT-4o-miniCategory=Closed-Source LLM2026.0435.2