--- title: "codebase-memory-mcp vs repobench" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/deusdata-codebase-memory-mcp-vs-leolty-repobench" tools: ["deusdata-codebase-memory-mcp", "leolty-repobench"] --- # codebase-memory-mcp vs repobench *GraphCanon updated Aug 25, 2026* ## Verdict Pick codebase-memory-mcp if codebase-memory-mcp excels at indexing codebases into a persistent knowledge graph with support for 158 languages, sub-ms query times, and minimal token use, delivered as a single static binary; pick repobench if repoBench assesses repository-level code auto-completion systems, offering benchmarks crucial for researchers and developers working on similar systems. [codebase-memory-mcp](https://deusdata.github.io/codebase-memory-mcp/) reports 41k GitHub stars, 3.3k forks, and 510 open issues, last pushed Aug 24, 2026. [repobench](https://arxiv.org/abs/2306.03091) has 214 stars, 13 forks, and 12 open issues, last pushed Aug 16, 2024. Figures are from public GitHub metadata via [codebase-memory-mcp's repository](https://github.com/DeusData/codebase-memory-mcp) and [repobench's repository](https://github.com/Leolty/repobench). | | [codebase-memory-mcp](/tools/deusdata-codebase-memory-mcp.md) | [repobench](/tools/leolty-repobench.md) | | --- | --- | --- | | Tagline | High-performance code intelligence MCP server indexing codebases into a persistent knowledge graph quickly | Benchmarking Repository-Level Code Auto-Completion Systems | | Stars | 40,581 | 214 | | Forks | 3,289 | 13 | | Open issues | 510 | 12 | | Language | C | Python | | Adopt for | codebase-memory-mcp excels at indexing codebases into a persistent knowledge graph with support for 158 languages, sub-ms query times, and minimal token use, delivered as a single static binary. | RepoBench assesses repository-level code auto-completion systems, offering benchmarks crucial for researchers and developers working on similar systems. | | Persona | - | - | | Runtime | - | - | | License | MIT | CC-BY-4.0 | | Categories | Data & Retrieval, Developer Tools | Developer Tools | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [codebase-memory-mcp](/tools/deusdata-codebase-memory-mcp.md) | [repobench](/tools/leolty-repobench.md) | | --- | --- | --- | | Maintenance | Very active (96%) | Dormant (18%) | | Days since push | 1d | 719d | | Open issues (now) | 510 | 12 | | Stars delta | +5.2k (30d) | Unknown | | Open issues delta | +162 (30d) | Unknown | | Full report | [trust report](/tools/deusdata-codebase-memory-mcp/trust.md) | [trust report](/tools/leolty-repobench/trust.md) | ## Decision facts: codebase-memory-mcp - **Adopt for:** codebase-memory-mcp excels at indexing codebases into a persistent knowledge graph with support for 158 languages, sub-ms query times, and minimal token use, delivered as a single static binary. ## Decision facts: repobench - **Adopt for:** RepoBench assesses repository-level code auto-completion systems, offering benchmarks crucial for researchers and developers working on similar systems. ## Choose when ### Choose codebase-memory-mcp if… - codebase-memory-mcp is primarily C; repobench is Python. - License: codebase-memory-mcp is MIT, repobench is CC-BY-4.0. - Tags unique to codebase-memory-mcp: aider, ast, claude-code, code-analysis. - Also covers Data & Retrieval. - When needing rapid codebase indexing into a knowledge graph with low latency queries ### Choose repobench if… - repobench is primarily Python; codebase-memory-mcp is C. - License: repobench is CC-BY-4.0, codebase-memory-mcp is MIT. - Tags unique to repobench: benchmarking, code-completion, iclr 2024. - Use RepoBench if you are evaluating the performance of your own repository-level code completion system against established benchmarks. ## When NOT to use codebase-memory-mcp - If you demand extensive interactive graphical UI features beyond its command-line interface capabilities - When your project prioritizes a dynamic runtime environment over static binaries ## When NOT to use repobench - Do not use RepoBench if your focus is on line-level or function-level code completions as opposed to repository-wide assessments. - Avoid RepoBench in scenarios where you require real-time analytics rather than benchmarking against static datasets and predefined criteria. ## Common questions ### What is the difference between codebase-memory-mcp and repobench? codebase-memory-mcp: High-performance code intelligence MCP server indexing codebases into a persistent knowledge graph quickly. repobench: Benchmarking Repository-Level Code Auto-Completion Systems. See the comparison table for live GitHub stats and shared categories. ### When should I choose codebase-memory-mcp over repobench? Choose codebase-memory-mcp over repobench when codebase-memory-mcp is primarily C; repobench is Python; License: codebase-memory-mcp is MIT, repobench is CC-BY-4.0; Tags unique to codebase-memory-mcp: aider, ast, claude-code, code-analysis; Also covers Data & Retrieval; When needing rapid codebase indexing into a knowledge graph with low latency queries. ### When should I choose repobench over codebase-memory-mcp? Choose repobench over codebase-memory-mcp when repobench is primarily Python; codebase-memory-mcp is C; License: repobench is CC-BY-4.0, codebase-memory-mcp is MIT; Tags unique to repobench: benchmarking, code-completion, iclr 2024; Use RepoBench if you are evaluating the performance of your own repository-level code completion system against established benchmarks. ### When should I avoid codebase-memory-mcp? If you demand extensive interactive graphical UI features beyond its command-line interface capabilities When your project prioritizes a dynamic runtime environment over static binaries ### When should I avoid repobench? Do not use RepoBench if your focus is on line-level or function-level code completions as opposed to repository-wide assessments. Avoid RepoBench in scenarios where you require real-time analytics rather than benchmarking against static datasets and predefined criteria. ### Is codebase-memory-mcp or repobench more popular on GitHub? codebase-memory-mcp has more GitHub stars (40,581 vs 214). Stars measure visibility, not whether either tool fits your constraints. ### Are codebase-memory-mcp and repobench open source? Yes - both are open-source projects on GitHub (codebase-memory-mcp: MIT, repobench: CC-BY-4.0). ### Where can I find alternatives to codebase-memory-mcp or repobench? GraphCanon lists graph-backed alternatives at [codebase-memory-mcp alternatives](/tools/deusdata-codebase-memory-mcp/alternatives) and [repobench alternatives](/tools/leolty-repobench/alternatives) ([codebase-memory-mcp markdown twin](/tools/deusdata-codebase-memory-mcp/alternatives.md), [repobench markdown twin](/tools/leolty-repobench/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/deusdata-codebase-memory-mcp-vs-leolty-repobench.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, codebase-memory-mcp or repobench? codebase-memory-mcp: Very active. repobench: 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 codebase-memory-mcp and repobench? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [codebase-memory-mcp trust report](/tools/deusdata-codebase-memory-mcp/trust); [repobench trust report](/tools/leolty-repobench/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=deusdata-codebase-memory-mcp`](/api/graphcanon/graph?tool=deusdata-codebase-memory-mcp) - 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/_