Skip to content

React Native Evals

An open benchmark for AI coding agents on real-world React Native implementation tasks, emphasizing working app behavior, recommended architecture choices, and strict constraint adherence.

15 models

Higher is better.

  1. 1Composer 2Cursor96.1
  2. 2GPT-5.4OpenAI85.3
  3. 3GPT-5.5OpenAI84.7
  4. 4Claude Opus 4.6Anthropic84.1
  5. 5Claude Opus 4.7Anthropic82.8
  6. 6Claude Sonnet 4.6Anthropic80.6
  7. 7Gemini 3.1 ProGoogle78.9
  8. 8Kimi K2.5Moonshot AI77.2
  9. 9Gemma 4 31BGoogle75.2
  10. 10GLM-5Z.AI74.8
  11. 11Grok 4xAI72.6
  12. 12GPT-OSS 120BOpenAI71.6
  13. 13DeepSeek V3.2DeepSeek71.5
  14. 14MiniMax M2.7MiniMax71.4
  15. 15GPT-OSS 20BOpenAI71.0
Higher is better. Bar colour: blue proprietary, orange open weights, green other.

About this benchmark

Metric
Framework-specific app development evaluation
Tasks
React Native app implementation tasks
Authors
Callstack (2026)

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.