T Thestageai Google OpenAI Xiaomi Head to head on 26 boards: OpenAI places best on 13 , Google places best on 10 , Xiaomi places best on 2 and Thestageai places best on 1 . All 4 ranked on 0 of 28 boards.
Board Read T Thestageai Google OpenAI Xiaomi Chat2 Text Arena Arena rating 1d yesterday Oct 6 1d yesterday #1 1,525 best of the picked Gemini 4 Argon #17 1,484 GPT-5.6 Sol #23 1,480 MiMo-V2.6-Pro MultiChallenge Score 1d yesterday Oct 6 1d yesterday #3 71.4% best of the picked Gemini 3.1 Pro Preview #4 69.2% GPT-5.4 Pro Reasoning7 AA Intelligence Index 1d yesterday Oct 6 1d yesterday #5 52.6 Gemini 4 Argon #4 52.7 best of the picked GPT-6 Astra #9 46.3 MiMo-V2.6-Pro HLE Accuracy 1d yesterday Oct 6 1d yesterday #3 46.4% Gemini 3.1 Pro Preview #1 54.8% best of the picked GPT-6 Astra GPQA Diamond Accuracy 1d yesterday Oct 6 1d yesterday #3 95.4% Gemini 3.8 Flash #1 95.8% best of the picked GPT-6 Astra ARC-AGI-2 Score 1d yesterday Oct 6 1d yesterday #8 89.2% Gemini 3.8 Flash #1 95.0% best of the picked GPT-6 Astra MMLU-Pro Accuracy · stale 1d yesterday Oct 6 1d yesterday #1 91.2% best of the picked Gemini 3.1 Pro Preview #3 89.3% o1 Epoch ECI ECI 1d yesterday Oct 6 1d yesterday #16 157.3 Gemini 3.7 Flash #2 166.4 best of the picked GPT-6 Astra LiveBench Global average 1d yesterday Oct 6 1d yesterday #14 78.8% Gemini 3.7 Flash #4 82.2% best of the picked GPT-6 Astra Coding4 WebDev Arena Arena rating 1d yesterday Oct 6 1d yesterday #8 1,680 Gemini 4 Argon #2 1,788 best of the picked GPT-6 Astra #20 1,618 MiMo-V2.6-Pro Aider Polyglot Pass rate · stale 1d yesterday Oct 6 1d yesterday #3 83.1% Gemini 2.5 Pro (Jun 2025) #1 88.0% best of the picked GPT-5 SWE-bench Verified Resolved · stale 1d yesterday Oct 6 1d yesterday #2 75.8% best of the picked Gemini 3 Flash #6 72.8% GPT-5.2 SWE-Bench Pro Resolved 1d yesterday Oct 6 1d yesterday #9 58.8% Gemini 3.8 Flash #3 90.2% best of the picked GPT-6 Astra Agents and tools3 Search Arena Arena rating 1d yesterday Oct 6 1d yesterday #9 1,210 Gemini 3.1 Pro Grounding #1 1,257 best of the picked GPT-5.6 Sol BFCL Overall accuracy 1d yesterday Oct 6 1d yesterday #3 72.5% best of the picked Gemini 3 Pro #7 63.0% o3 MCP Atlas Pass rate 1d yesterday Oct 6 1d yesterday #6 83.6% best of the picked Gemini 3.5 Flash #11 81.8% GPT-5.6 Vision2 Vision Arena Arena rating 1d yesterday Oct 6 1d yesterday #5 1,296 best of the picked Gemini 3.7 Flash #8 1,291 GPT-6.1 Sol #46 1,247 MiMo-V2.6-Flash Document Arena Arena rating 1d yesterday Oct 6 1d yesterday #16 1,463 Gemini 3.5 Flash #6 1,486 best of the picked GPT-5.5 Image2 Text-to-Image Arena rating 1d yesterday Oct 6 1d yesterday #9 1,261 Gemini 3.1 Flash Image #1 1,425 best of the picked GPT Image 2.5 Sunburst Image Edit Arena rating 1d yesterday Oct 6 1d yesterday #10 1,390 Gemini 3 Pro Image #1 1,524 best of the picked GPT Image 2.5 Sunburst Video2 Text-to-Video Arena rating 1d yesterday Oct 6 1d yesterday #1 1,516 best of the picked Gemini Omni 1.1 Flash #11 1,368 Sora 2 Pro Image-to-Video Arena rating 1d yesterday Oct 6 1d yesterday #2 1,488 Gemini Omni 1.1 Flash Speech2 Speech Arena Arena rating 1d yesterday Oct 6 1d yesterday #5 1,275 best of the picked Gemini 3.8 Flash TTS #36 1,103 TTS 1 Hd #59 1,055 MiMo V2.5 TTS Open ASR Average WER 1d yesterday Oct 6 1d yesterday #18 4.54 WER best of the picked thewhisper-large-v3-turbo (TheStageAI) #62 9.51 WER gemma-4-E4B-it (google) #20 4.57 WER gpt-transcribe (openai) Embeddings1 MTEB Mean task score 1d yesterday Oct 6 1d yesterday #5 68.4% gemini-embedding-001 (google) Speed1 AA Speed Output speed 1d yesterday Oct 6 1d yesterday #1 249 tok/s best of the picked Gemini 3.8 Flash #6 138 tok/s GPT-6 Luna #20 51 tok/s MiMo-V2.6-Flash Price1 AA Price Blended price 1d yesterday Oct 6 1d yesterday #10 $1.50 Gemini 3.8 Flash #2 $0.20 GPT-6 Luna #1 $0.17 best of the picked MiMo-V2.6-Flash Usage1 OpenRouter Weekly tokens 1d yesterday Oct 6 1d yesterday #13 2.2T tok Gemini 3.8 Flash #6 6.2T tok GPT-6 Luna #4 9.7T tok best of the picked MiMo-V2.6-Flash
Green marks the best placing among the picked where two or more are ranked; ties are marked for each. Ranks are each board's own order (lower is better), scores as the board published them. Dash: not on that board.