Frontier-CS: Benchmarking LLMs on Computer Science Problems

Active

238 open-ended computer science problems spanning algorithmic (172) and research (66) tracks. Problems feature continuous partial scoring, with algorithmic solutions evaluated via compilation and test-case checking,

Open

Publisher: University of California, Berkeley
Domain: Coding
License: mit
Published: Mar 2026
Notable for: Benchmark for evaluating Coding.
Canonical: github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/frontier_cs

Cite

Notes

Only stored in your browser.

Attribution

README: github.com/UKGovernmentBEIS/inspect_evals/blob/main/src/inspect_evals/frontier_cs/eval.yamlMIT

Attribution policy →

FAQ

What is Frontier-CS: Benchmarking LLMs on Computer Science Problems?: 238 open-ended computer science problems spanning algorithmic (172) and research (66) tracks. Problems feature continuous partial scoring, with algorithmic solutions evaluated via compilation and test-case checking,
What license is Frontier-CS: Benchmarking LLMs on Computer Science Problems under?: Frontier-CS: Benchmarking LLMs on Computer Science Problems is available under mit.