OpenResearch

AI model rankings for autoresearch

Live LLM rankings by real-world autoresearch usage on OpenResearch.

Claude Opus 5.5 · 38.0% · AnthropicClaude Opus 5.5
38.0%
GPT-6.1 Sol · 15.0% · OpenAIGPT-6.1 Sol
15.0%
Other Models · 7.9% · OtherOther Models
7.9%
GPT-6 Astra · 6.8% · OpenAIGPT-6 Astra
6.8%
DeepSeek V4.1 Flash · 6.0% · DeepSeekDeepSeek V4.1 Flash
6.0%
Space Bunny · 4.9% · OpenCodeSpace Bunny
4.9%
GPT-6 Luna · 4.7% · OpenAIGPT-6 Luna
4.7%
Muse Spark 1.3 · 4.7% · MetaMuse Spark 1.3
4.7%
Big Pickle · 3.4% · OpenCodeBig Pickle
3.4%
Claude Fable 5.1 · 3.2% · AnthropicClaude Fable 5.1
3.2%
GPT-6 Sol · 2.8% · OpenAIGPT-6 Sol
2.8%
Claude Sonnet 5.5 · 2.5% · AnthropicClaude Sonnet 5.5
2.5%
AnthropicOpenAIDeepSeekOpenCodeMetaOther
  1. Claude Opus 5.5: 38.0%
  2. GPT-6.1 Sol: 15.0%
  3. GPT-6 Astra: 6.8%
  4. DeepSeek V4.1 Flash: 6.0%
  5. Space Bunny: 4.9%
  6. GPT-6 Luna: 4.7%
  7. Muse Spark 1.3: 4.7%
  8. Big Pickle: 3.4%
  9. Claude Fable 5.1: 3.2%
  10. GPT-6 Sol: 2.8%
  11. Claude Sonnet 5.5: 2.5%
  12. Other Models: 7.9%

Tile area represents each model’s share of tokens. Data collected since October 2, 2026.