Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-turn task completion on FunctionGym
Loading...
42.3
Task Completion Score
GRPO + Qwen2.5-8B
6.316
15.658
25
34.342
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
42.3
Gemini-1.5-Pro
Category=Closed-Source...
2026.04
41
GRPO + Qwen2.5-4B
Category=RL (GRPO) Tra...
2026.04
39.7
Gemini-1.5-Flash
Category=Closed-Source...
2026.04
32.1
Qwen2.5-14B + PRIME
Category=PRIME Models
2026.04
29.8
GPT-4o
Category=Closed-Source...
2026.04
28.2
Qwen2.5-8B + PRIME
Category=PRIME Models
2026.04
27.8
Qwen2.5-4B + PRIME
Category=PRIME Models
2026.04
24.5
Qwen2.5-14B
Category=Open-Source LLM
2026.04
16.7
GPT-4o-mini
Category=Closed-Source...
2026.04
15.4
Qwen2.5-32B
Category=Open-Source LLM
2026.04
15.4
Qwen2.5-8B
Category=Open-Source LLM
2026.04
9.5
Qwen2.5-4B
Category=Open-Source LLM
2026.04
7.7
Feedback
Search any
task
Search any
task