Home Compare Compare runs Pick up to 4 runs - the selection is encoded in the URL, so the comparison is shareable and bookmarkable.
Gemini Flash 3.5 + Gemini Flash Lite (circles, 81.8) Mimo v2.5 (circles, 81.2) Muse Spark 1.1 (circles, 80.7) Plus qwen3.7 (circles, 79.9) Gemini Flash 3.5 + Gemini Flash Lite (circles, 79.6) Gpt Terra 5.6 + 6 others (circles, 79.0) Gemini Flash Lite 3.5 (circles, 78.2) Plus qwen3.7 (circles, 76.7) Mimo v2.5 (solarSystem, 76.6) Flash qwen3.7 (circles, 76.5) Solar pro4 (circles, 75.9) Auto (circles, 75.7) Gemini Flash 3.5 + Gemini Flash Lite (circles, 74.7) Gemini Flash 3.6 (circles, 73.1) Gemini Flash 3.6 + Gemini Flash Lite (circles, 72.0) Claude Sonnet 5 (circles, 70.8) Deepseek Flash v4 (circles, 70.2) Muse Spark 1.2 + 2 others (arenaEvolution, 70.0) Muse Spark 1.1 (solarSystem, 67.8) Mimo Pro v2.5 + Mimo (arenaEvolution, 67.8) Gemini Flash Lite 3.1 (circles, 65.7) Gemini Flash 3.5 + Gemini Flash Lite (circles, 65.5) Mimo v2.5 (circles, 65.0) Gemini Flash 3.6 + Gemini Flash Lite (solarSystem, 64.6) Grok 4.5 (circles, 64.6) Mimo Pro v2.5 (circles, 64.4) Mimo Pro v2.5 + Mimo (circles, 64.3) Plus qwen3.7 (circles, 64.2) Gemini Flash Lite 3.1 (circles, 64.1) Deepseek Flash v4 (circles, 64.0) Gemini Flash Lite 3.1 (solarSystem, 63.9) Mimo Pro v2.5 + Mimo (solarSystem, 63.9) Mimo Pro v2.5 + Mimo (the2048, 63.3) Kimi k2.6 (circles, 63.2) Grok 4.5 + Deepseek Flash (the2048, 63.2) Grok 4.5 (circles, 63.1) Mimo Pro v2.5 (circles, 63.0) Gpt Terra 5.6 + 6 others (solarSystem, 62.8) Grok 4.5 (circles, 62.2) Glm 5.2 (circles, 61.6) Gpt Terra 5.6 (circles, 61.4) Gemini Flash 3.6 (circles, 61.0) Kimi Code k2.7 (circles, 60.9) Gpt Luna 5.6 (circles, 60.8) Grok 4.5 (circles, 60.7) Mimo Pro v2.5 (solarSystem, 60.2) Muse Spark 1.2 + Gemini Flash Lite (the2048, 59.6) Gemini Flash 3.5 + Gemini Flash Lite (solarSystem, 59.3) Step Flash 3.7 (circles, 59.1) Kimi Code k2.7 (circles, 59.1) Claude Sonnet 5 + Gemini Flash Lite (solarSystem, 58.9) Deepseek Flash v4 (the2048, 58.5) Muse Spark 1.2 + 2 others (solarSystem, 58.0) Claude Sonnet 5 (circles, 57.9) Step Flash 3.7 (solarSystem, 57.8) Deepseek Flash v4 (circles, 57.1) Muse Spark 1.1 (the2048, 56.8) Gemini Flash 3 (preview) (circles, 56.4) Gemini Flash 3 (preview) (circles, 56.4) Deepseek Flash v4 (circles, 56.1) Plus qwen3.7 (solarSystem, 55.7) Deepseek Pro v4 (circles, 55.6) Glm 5.2 (circles, 54.2) Gemini Flash 3 (preview) (circles, 54.2) Gemini Flash 3.6 + Gemini Flash Lite (the2048, 53.6) Deepseek Pro v4 (circles, 53.6) Gemini Flash 3 (preview) (circles, 53.6) Gemini Pro 3.1 (preview) (circles, 53.2) Kimi k2.6 (circles, 52.8) Gpt Terra 5.6 + Gpt Luna (the2048, 52.5) Deepseek Flash v4 (solarSystem, 52.5) Gemini Flash 3.5 (circles, 52.3) Gemini Pro 3.1 (preview) (circles, 52.0) Gpt Terra 5.6 + 6 others (the2048, 51.9) Inkling (circles, 51.7) Deepseek Pro v4 (circles, 51.6) Gemini Flash Lite 3.1 (circles, 51.0) Kimi k2.6 (circles, 49.7) Inkling (circles, 48.6) Claude Sonnet 5 (circles, 47.8) Gemini Flash 3.5 (circles, 47.3) Glm 5.2 (circles, 46.7) Claude Sonnet 5 (circles, 46.5) Gemini Flash 3.5 + Gemini Flash Lite (solarSystem, 44.6) Claude Sonnet 4.6 (circles, 44.6) Claude Sonnet 4.6 (circles, 42.5) Gemini Flash 3 (preview) (circles, 38.9) Gemini Flash 3 (preview) (solarSystem, 30.6)Select at least two runs to compare.
Every run on Favur Evals is scored by the same deterministic engine, on the same Statements of Work. The benchmark is self-funded - no vendor sponsorship, credits, or grants.