--- title: "FastChat vs evals" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/lm-sys-fastchat-vs-openai-evals" tools: ["lm-sys-fastchat", "openai-evals"] --- # FastChat vs evals *GraphCanon updated Sep 20, 2026* ## Verdict Pick FastChat if fastChat is an open-source platform for training, serving, and evaluating large language models, including Vicuna, and supports Chatbot Arena, a platform for LLM evaluations with human feedback; pick evals if evals is an evaluation framework from OpenAI for assessing large language models and systems built with them. It includes an open-source registry of benchmarks and tools to. [FastChat](https://github.com/lm-sys/FastChat) reports 40k GitHub stars, 4.8k forks, and 1.0k open issues, last pushed May 1, 2026. [evals](https://github.com/openai/evals) has 19k stars, 3.1k forks, and 333 open issues, last pushed Apr 14, 2026. Figures are from public GitHub metadata via [FastChat's repository](https://github.com/lm-sys/FastChat) and [evals's repository](https://github.com/openai/evals). | | [FastChat](/tools/lm-sys-fastchat.md) | [evals](/tools/openai-evals.md) | | --- | --- | --- | | Tagline | An open platform for training, serving, and evaluating large language models | Framework for evaluating LLMs and LLM systems with an open-source registry of benchmarks. | | Stars | 39,536 | 19,398 | | Forks | 4,774 | 3,077 | | Open issues | 1,043 | 333 | | Language | Python | Python | | Adopt for | FastChat is an open-source platform for training, serving, and evaluating large language models, including Vicuna, and supports Chatbot Arena, a platform for LLM evaluations with human feedback. | Evals is an evaluation framework from OpenAI for assessing large language models and systems built with them. It includes an open-source registry of benchmarks and tools to create custom evaluations. | | Persona | - | - | | Runtime | - | - | | License | Apache-2.0 | Other | | Categories | Evaluation & Observability, Inference & Serving, LLM Frameworks, Model Training | Evaluation & Observability | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [FastChat](/tools/lm-sys-fastchat.md) | [evals](/tools/openai-evals.md) | | --- | --- | --- | | Days since push | 140d | 145d | | Open issues (now) | 1.0k | 333 | | Stars delta | +19 (30d) | +271 (30d) | | Open issues delta | +5 (30d) | +120 (30d) | | Full report | [trust report](/tools/lm-sys-fastchat/trust.md) | [trust report](/tools/openai-evals/trust.md) | ## Shared compatibility - **Python**: [FastChat](/tools/lm-sys-fastchat.md) - Python runtime; [evals](/tools/openai-evals.md) - Python runtime ## Decision facts: FastChat - **Adopt for:** FastChat is an open-source platform for training, serving, and evaluating large language models, including Vicuna, and supports Chatbot Arena, a platform for LLM evaluations with human feedback. ## Decision facts: evals - **Adopt for:** Evals is an evaluation framework from OpenAI for assessing large language models and systems built with them. It includes an open-source registry of benchmarks and tools to create custom evaluations. ## Choose when ### Choose FastChat if… - License: FastChat is Apache-2.0, evals is Other. - Tags unique to FastChat: chatbot arena, llm-evaluation, multi-model serving, openai-compatible apis. - Also covers Inference & Serving, LLM Frameworks, Model Training. - When you need a platform that supports both training and evaluation of large language models, such as Vicuna, and you want to leverage human feedback for model evaluation. ### Choose evals if… - License: evals is Other, FastChat is Apache-2.0. - Tags unique to evals: benchmarking, custom eval creation, evaluation-framework, large-language-models. - * When you need a comprehensive set of pre-existing evals and the ability to create your own tailored tests using specific use cases, especially within the OpenAI model ecosystem. ## When NOT to use FastChat - If you are looking for a platform that focuses solely on model training without serving or evaluation capabilities, FastChat offers a comprehensive suite that might be more than you need. - If your project does not require human feedback for model evaluation, FastChat's integration with Chatbot Arena might be an unnecessary feature. - If you are working with proprietary or closed-source models, FastChat's open-source nature and reliance on open-source models like Vicuna might not align with your project requirements. ## When NOT to use evals - * When evaluating models or systems that do not benefit from being integrated with the OpenAI API, as some features like direct evals configuration in the OpenAI Dashboard require an OpenAI key. - * If you are looking for an evaluation framework that doesn’t involve external dependencies such as Git Large File Storage (LFS) and specific Python version requirements (Python 3.9 minimum), or if a ## Common questions ### What is the difference between FastChat and evals? FastChat: An open platform for training, serving, and evaluating large language models. evals: Framework for evaluating LLMs and LLM systems with an open-source registry of benchmarks.. See the comparison table for live GitHub stats and shared categories. ### When should I choose FastChat over evals? Choose FastChat over evals when License: FastChat is Apache-2.0, evals is Other; Tags unique to FastChat: chatbot arena, llm-evaluation, multi-model serving, openai-compatible apis; Also covers Inference & Serving, LLM Frameworks, Model Training; When you need a platform that supports both training and evaluation of large language models, such as Vicuna, and you want to leverage human feedback for model evaluation. ### When should I choose evals over FastChat? Choose evals over FastChat when License: evals is Other, FastChat is Apache-2.0; Tags unique to evals: benchmarking, custom eval creation, evaluation-framework, large-language-models; * When you need a comprehensive set of pre-existing evals and the ability to create your own tailored tests using specific use cases, especially within the OpenAI model ecosystem. ### When should I avoid FastChat? If you are looking for a platform that focuses solely on model training without serving or evaluation capabilities, FastChat offers a comprehensive suite that might be more than you need. If your project does not require human feedback for model evaluation, FastChat's integration with Chatbot Arena might be an unnecessary feature. If you are working with proprietary or closed-source models, FastChat's open-source nature and reliance on open-source models like Vicuna might not align with your project requirements. ### When should I avoid evals? * When evaluating models or systems that do not benefit from being integrated with the OpenAI API, as some features like direct evals configuration in the OpenAI Dashboard require an OpenAI key. * If you are looking for an evaluation framework that doesn’t involve external dependencies such as Git Large File Storage (LFS) and specific Python version requirements (Python 3.9 minimum), or if a ### Is FastChat or evals more popular on GitHub? FastChat has more GitHub stars (39,536 vs 19,398). Stars measure visibility, not whether either tool fits your constraints. ### Are FastChat and evals open source? Yes - both are open-source projects on GitHub (FastChat: Apache-2.0, evals: Other). ### Where can I find alternatives to FastChat or evals? GraphCanon lists graph-backed alternatives at [FastChat alternatives](/tools/lm-sys-fastchat/alternatives) and [evals alternatives](/tools/openai-evals/alternatives) ([FastChat markdown twin](/tools/lm-sys-fastchat/alternatives.md), [evals markdown twin](/tools/openai-evals/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/lm-sys-fastchat-vs-openai-evals.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, FastChat or evals? FastChat: Slowing. evals: Slowing. 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 FastChat and evals? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [FastChat trust report](/tools/lm-sys-fastchat/trust); [evals trust report](/tools/openai-evals/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=lm-sys-fastchat`](/api/graphcanon/graph?tool=lm-sys-fastchat) - 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/_