--- title: "code-eval vs lever" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/abacaj-code-eval-vs-niansong1996-lever" tools: ["abacaj-code-eval", "niansong1996-lever"] --- # code-eval vs lever *GraphCanon updated Oct 7, 2026* ## Verdict Pick code-eval if code-eval assesses large language models with the human-eval benchmark to provide insights into code generation reliability; pick lever if lever offers support for verifying language-to-code generation through actual code execution. [code-eval](https://github.com/abacaj/code-eval) reports 432 GitHub stars, 36 forks, and 5 open issues, last pushed Sep 12, 2023. [lever](https://arxiv.org/abs/2302.08468) has 89 stars, 8 forks, and 2 open issues, last pushed Jul 5, 2023. Figures are from public GitHub metadata via [code-eval's repository](https://github.com/abacaj/code-eval) and [lever's repository](https://github.com/niansong1996/lever). | | [code-eval](/tools/abacaj-code-eval.md) | [lever](/tools/niansong1996-lever.md) | | --- | --- | --- | | Tagline | Run evaluation on LLMs using human-eval benchmark. | Supports learning to verify language-to-code generation with execution | | Stars | 432 | 89 | | Forks | 36 | 8 | | Open issues | 5 | 2 | | Language | Python | Python | | Adopt for | code-eval assesses large language models with the human-eval benchmark to provide insights into code generation reliability. | Lever offers support for verifying language-to-code generation through actual code execution. | | Persona | - | - | | Runtime | - | - | | License | MIT | Lever's source code is freely available under an MIT License for modification and distribution in both personal and commercial projects. | | Categories | Evaluation & Observability | Evaluation & Observability, Model Training | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [code-eval](/tools/abacaj-code-eval.md) | [lever](/tools/niansong1996-lever.md) | | --- | --- | --- | | Days since push | 1120d | 1189d | | Open issues (now) | 5 | 2 | | Stars delta | +1 (30d) | -1 (30d) | | Full report | [trust report](/tools/abacaj-code-eval/trust.md) | [trust report](/tools/niansong1996-lever/trust.md) | ## Shared compatibility - **Python**: [code-eval](/tools/abacaj-code-eval.md) - Python runtime; [lever](/tools/niansong1996-lever.md) - Python runtime ## Decision facts: code-eval - **Adopt for:** code-eval assesses large language models with the human-eval benchmark to provide insights into code generation reliability. ## Decision facts: lever - **Requirements:** All of the pipelines have been tested on Linux machines only, requiring possibly custom `tree-sitter` parsers for other platforms. - **Adopt for:** Lever offers support for verifying language-to-code generation through actual code execution. - **License detail:** Lever's source code is freely available under an MIT License for modification and distribution in both personal and commercial projects. ## Choose when ### Choose code-eval if… - Tags unique to code-eval: humaneval, wizardcoder. - When you need clear comparisons of pass rates for different LLMs using standardized tests - More GitHub stars (432 vs 89) - visibility, not fit. ### Choose lever if… - Requirements: All of the pipelines have been tested on Linux machines only, requiring possibly custom `tree-sitter` parsers for other platforms.. - Tags unique to lever: code verification, execution based verification, language-to-code. - Also covers Model Training. - When the development team needs to verify accuracy of generated code from language inputs based on execution, and has access to Linux machines to ensure seamless functionality. ## When NOT to use code-eval - If the tool's results do not correlate well with the official published benchmarks due to unknown prompt differences - For real-time or dynamic evaluations as this repo offers pre-computed static results only ## When NOT to use lever - Avoid Lever if developing in an environment other than Linux as it requires the use of tree-sitter parsers, which may not be compatible with your system. - Do not choose Lever if you seek a tool that does not require setting up a conda environment and installing specific dependencies to operate. ## Common questions ### What is the difference between code-eval and lever? code-eval: Run evaluation on LLMs using human-eval benchmark.. lever: Supports learning to verify language-to-code generation with execution. See the comparison table for live GitHub stats and shared categories. ### When should I choose code-eval over lever? Choose code-eval over lever when Tags unique to code-eval: humaneval, wizardcoder; When you need clear comparisons of pass rates for different LLMs using standardized tests; More GitHub stars (432 vs 89) - visibility, not fit. ### When should I choose lever over code-eval? Choose lever over code-eval when Requirements: All of the pipelines have been tested on Linux machines only, requiring possibly custom `tree-sitter` parsers for other platforms.; Tags unique to lever: code verification, execution based verification, language-to-code; Also covers Model Training; When the development team needs to verify accuracy of generated code from language inputs based on execution, and has access to Linux machines to ensure seamless functionality. ### When should I avoid code-eval? If the tool's results do not correlate well with the official published benchmarks due to unknown prompt differences For real-time or dynamic evaluations as this repo offers pre-computed static results only ### When should I avoid lever? Avoid Lever if developing in an environment other than Linux as it requires the use of tree-sitter parsers, which may not be compatible with your system. Do not choose Lever if you seek a tool that does not require setting up a conda environment and installing specific dependencies to operate. ### Is code-eval or lever more popular on GitHub? code-eval has more GitHub stars (432 vs 89). Stars measure visibility, not whether either tool fits your constraints. ### Are code-eval and lever open source? Yes - both are open-source projects on GitHub (code-eval: MIT, lever: MIT). ### Where can I find alternatives to code-eval or lever? GraphCanon lists graph-backed alternatives at [code-eval alternatives](/tools/abacaj-code-eval/alternatives) and [lever alternatives](/tools/niansong1996-lever/alternatives) ([code-eval markdown twin](/tools/abacaj-code-eval/alternatives.md), [lever markdown twin](/tools/niansong1996-lever/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/abacaj-code-eval-vs-niansong1996-lever.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, code-eval or lever? code-eval: Dormant. lever: Dormant. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for code-eval and lever? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [code-eval trust report](/tools/abacaj-code-eval/trust); [lever trust report](/tools/niansong1996-lever/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=abacaj-code-eval`](/api/graphcanon/graph?tool=abacaj-code-eval) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_