Scicode Zero Shot Agent

Agent performance overview across all HAL benchmarks

1
Benchmarks
9
Models Used
2
Pareto Optimal Runs

Benchmark Performance

On the Pareto Frontier? indicates whether this agent achieved a Pareto-optimal trade-off between accuracy and cost on that benchmark. Agents on the Pareto frontier represent the current state-of-the-art efficiency for their performance level.

Benchmark Model Accuracy Cost On the Pareto Frontier?
Scicode
o4-mini Low (April 2025) 9.23% $1.74 Yes
Scicode
GPT-4.1 (April 2025) 6.15% $2.82 No
Scicode
o4-mini High (April 2025) 6.15% $5.37 No
Scicode
o3 Medium (April 2025) 4.62% $6.03 No
Scicode
DeepSeek V3 (March 2025) 3.08% $0.79 No
Scicode
Claude-3.7 Sonnet High (February 2025) 3.08% $4.99 No
Scicode
Gemini 2.0 Flash (February 2025) 1.54% $0.12 Yes
Scicode
DeepSeek R1 (May 2025) 0.00% $2.19 No
Scicode
Claude-3.7 Sonnet (February 2025) 0.00% $5.10 No