Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Negotiation on AmazonHistoryPrice (gpt-5.4-high-reasoning seller, test)
Loading...
0.408
Reward
Qwen3-30B-A3B-Instruct-2507-trained
-0.30128
-0.11714
0.067
0.25114
Apr 10, 2026
Reward
Deal Rate
Bargained Ratio
Updated 5d ago
Evaluation Results
Method
Method
Links
Reward
Deal Rate
Bargained Ratio
Qwen3-30B-A3B-Instruct-2507-trained
Training status=trained
2026.04
0.408
75
40.8
gpt-5.4-high-reasoning
Reasoning configuratio...
2026.04
0.182
91.4
18.2
gpt-5.4-no-reasoning
Reasoning configuratio...
2026.04
0.146
84.8
16.1
DeepSeek-V3.1-thinking
Reasoning configuratio...
2026.04
0.122
92.6
13
DeepSeek-V3.1-nothink
Reasoning configuratio...
2026.04
0.12
90.6
14.8
Kimi-K2-Thinking
Reasoning configuratio...
2026.04
0.106
90.6
12.6
Qwen3-30B-A3B-Instruct-2507-untrained
Training status=untrained
2026.04
-0.274
60.5
3.9
Feedback
Search any
task
Search any
task