Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-turn task completion on TauGym
Loading...
21
Task Completion Score
GRPO + Qwen2.5-8B
-0.84
4.83
10.5
16.17
Apr 8, 2026
Task Completion Score
Updated 3mo ago
Evaluation Results
Method
Method
Links
Task Completion Score
GRPO + Qwen2.5-8B
Category=RL (GRPO) Tra...
2026.04
21
GPT-4o-mini
Category=Closed-Source...
2026.04
20.6
GRPO + Qwen2.5-4B
Category=RL (GRPO) Tra...
2026.04
20
Gemini-1.5-Pro
Category=Closed-Source...
2026.04
19.4
Qwen2.5-14B + PRIME
Category=PRIME Models
2026.04
12.4
Gemini-1.5-Flash
Category=Closed-Source...
2026.04
12.1
Qwen2.5-14B
Category=Open-Source LLM
2026.04
10.3
Qwen2.5-8B + PRIME
Category=PRIME Models
2026.04
7.2
Qwen2.5-4B + PRIME
Category=PRIME Models
2026.04
5.8
Qwen2.5-8B
Category=Open-Source LLM
2026.04
4.8
Qwen2.5-4B
Category=Open-Source LLM
2026.04
3.6
GPT-4o
Category=Closed-Source...
2026.04
3
Qwen2.5-32B
Category=Open-Source LLM
2026.04
0
Feedback
Search any
task
Search any
task