Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-horizon game playing on BALROG MiniHack
Loading...
40
Progression Rate
Gemini-3-Pro
-1.6
9.2
20
30.8
Jul 1, 2026
Progression Rate
Updated 24d ago
Evaluation Results
Method
Method
Links
Progression Rate
Gemini-3-Pro
category=Frontier prop...
2026.07
40
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
30
Gemini-3.1-Pro-Thinking
category=Frontier prop...
2026.07
27.5
Claude-Opus-4.5
category=Frontier prop...
2026.07
27.5
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
27.5
DeepSeek-R1
category=Open-weight (...
2026.07
25
Gemini-2.5-Pro
category=Frontier prop...
2026.07
17.5
Qwen2.5-32B-Instruct
context-management met...
2026.07
10
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
7.5
Qwen2.5-72B-Instruct
category=Open-weight (...
2026.07
5
Qwen2.5-32B-Instruct
context-management met...
2026.07
2.5
Qwen2.5-7B-Instruct
category=Open-weight (...
2026.07
0
Feedback
Search any
task
Search any
task