--- title: "R2R vs SAG" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/sciphi-ai-r2r-vs-zleap-ai-sag" tools: ["sciphi-ai-r2r", "zleap-ai-sag"] --- # R2R vs SAG *GraphCanon updated Aug 23, 2026* ## Verdict Pick R2R if r2R is a state-of-the-art AI retrieval system that's production-ready and can easily be integrated through its RESTful API, employing Agentic Retrieval-Augmented Generation (RAG); pick SAG if sAG is a document retrieval project built with TypeScript to aid in efficient search and retrieval within knowledge bases. [R2R](https://github.com/SciPhi-AI/R2R) reports 8.0k GitHub stars, 645 forks, and 122 open issues, last pushed Nov 7, 2025. [SAG](https://zleap.com) has 2.4k stars, 148 forks, and 2 open issues, last pushed Aug 22, 2026. Figures are from public GitHub metadata via [R2R's repository](https://github.com/SciPhi-AI/R2R) and [SAG's repository](https://github.com/Zleap-AI/SAG). | | [R2R](/tools/sciphi-ai-r2r.md) | [SAG](/tools/zleap-ai-sag.md) | | --- | --- | --- | | Tagline | SoTA production-ready AI retrieval system with RESTful API | Document retrieval system built on SAG | | Stars | 7,967 | 2,406 | | Forks | 645 | 148 | | Open issues | 122 | 2 | | Language | Python | TypeScript | | Adopt for | R2R is a state-of-the-art AI retrieval system that's production-ready and can easily be integrated through its RESTful API, employing Agentic Retrieval-Augmented Generation (RAG). | SAG is a document retrieval project built with TypeScript to aid in efficient search and retrieval within knowledge bases. | | Persona | - | - | | Runtime | - | - | | License | MIT | MIT | | Categories | Data & Retrieval, Inference & Serving | AI Agents, Data & Retrieval | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [R2R](/tools/sciphi-ai-r2r.md) | [SAG](/tools/zleap-ai-sag.md) | | --- | --- | --- | | Maintenance | Slowing (36%) | Very active (96%) | | Days since push | 283d | 0d | | Open issues (now) | 122 | 2 | | Stars delta | +36 (30d) | +190 (30d) | | Open issues delta | 0 (30d) | +2 (30d) | | Full report | [trust report](/tools/sciphi-ai-r2r/trust.md) | [trust report](/tools/zleap-ai-sag/trust.md) | ## Shared compatibility - **Python**: [R2R](/tools/sciphi-ai-r2r.md) - Python runtime; [SAG](/tools/zleap-ai-sag.md) - Python runtime ## Decision facts: R2R - **Adopt for:** R2R is a state-of-the-art AI retrieval system that's production-ready and can easily be integrated through its RESTful API, employing Agentic Retrieval-Augmented Generation (RAG). ## Decision facts: SAG - **Adopt for:** SAG is a document retrieval project built with TypeScript to aid in efficient search and retrieval within knowledge bases. ## Choose when ### Choose R2R if… - R2R is primarily Python; SAG is TypeScript. - Tags unique to R2R: artificial-intelligence, large language models, python, question-answering. - Also covers Inference & Serving. - When you require top-tier accuracy in an AI-based retrieval system with the ease of integration provided by a RESTful API. ### Choose SAG if… - SAG is primarily TypeScript; R2R is Python. - Tags unique to SAG: agent, ai, data-engineering, knowledge-graph. - Also covers AI Agents. - When you need graph and vector-based techniques for retrieving documents ## When NOT to use R2R - If the application does not benefit from or requires less sophisticated methods of data retrieval that do not include RAG. - When integrating with systems or in environments where network latency might impede performance due to its dependency on a RESTful API interface. ## When NOT to use SAG - Avoid if the project requires features not supported by TypeScript, favoring alternative languages or environments instead - Do not use SAG when the architecture of your system cannot benefit from graph and vector-based retrieval methods, as it may lead to underutilization of its capabilities ## Common questions ### What is the difference between R2R and SAG? R2R: SoTA production-ready AI retrieval system with RESTful API. SAG: Document retrieval system built on SAG. See the comparison table for live GitHub stats and shared categories. ### When should I choose R2R over SAG? Choose R2R over SAG when R2R is primarily Python; SAG is TypeScript; Tags unique to R2R: artificial-intelligence, large language models, python, question-answering; Also covers Inference & Serving; When you require top-tier accuracy in an AI-based retrieval system with the ease of integration provided by a RESTful API. ### When should I choose SAG over R2R? Choose SAG over R2R when SAG is primarily TypeScript; R2R is Python; Tags unique to SAG: agent, ai, data-engineering, knowledge-graph; Also covers AI Agents; When you need graph and vector-based techniques for retrieving documents. ### When should I avoid R2R? If the application does not benefit from or requires less sophisticated methods of data retrieval that do not include RAG. When integrating with systems or in environments where network latency might impede performance due to its dependency on a RESTful API interface. ### When should I avoid SAG? Avoid if the project requires features not supported by TypeScript, favoring alternative languages or environments instead Do not use SAG when the architecture of your system cannot benefit from graph and vector-based retrieval methods, as it may lead to underutilization of its capabilities ### Is R2R or SAG more popular on GitHub? R2R has more GitHub stars (7,967 vs 2,406). Stars measure visibility, not whether either tool fits your constraints. ### Are R2R and SAG open source? Yes - both are open-source projects on GitHub (R2R: MIT, SAG: MIT). ### Where can I find alternatives to R2R or SAG? GraphCanon lists graph-backed alternatives at [R2R alternatives](/tools/sciphi-ai-r2r/alternatives) and [SAG alternatives](/tools/zleap-ai-sag/alternatives) ([R2R markdown twin](/tools/sciphi-ai-r2r/alternatives.md), [SAG markdown twin](/tools/zleap-ai-sag/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/sciphi-ai-r2r-vs-zleap-ai-sag.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, R2R or SAG? R2R: Slowing. SAG: Very active. 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 R2R and SAG? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [R2R trust report](/tools/sciphi-ai-r2r/trust); [SAG trust report](/tools/zleap-ai-sag/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=sciphi-ai-r2r`](/api/graphcanon/graph?tool=sciphi-ai-r2r) - 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/_