Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-horizon game playing on BALROG NetHack
Loading...
6.8
Progression Rate
Gemini-3-Pro
-0.272
1.564
3.4
5.236
Jul 1, 2026
Progression Rate
Updated 24d ago
Evaluation Results
Method
Method
Links
Progression Rate
Gemini-3-Pro
category=Frontier prop...
2026.07
6.8
Gemini-3.1-Pro-Thinking
category=Frontier prop...
2026.07
2.6
Claude-Opus-4.5
category=Frontier prop...
2026.07
2
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
1.85
Gemini-2.5-Pro
category=Frontier prop...
2026.07
1.7
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
1.57
DeepSeek-R1
category=Open-weight (...
2026.07
1.4
AUTOMEM
backbone=Qwen2.5-32B-I...
2026.07
0.42
Qwen2.5-72B-Instruct
category=Open-weight (...
2026.07
0.3
Qwen2.5-7B-Instruct
category=Open-weight (...
2026.07
0
Qwen2.5-32B-Instruct
context-management met...
2026.07
0
Qwen2.5-32B-Instruct
context-management met...
2026.07
0
Feedback
Search any
task
Search any
task