Moonshot AI Anthropic Google Head to head on 21 boards: Anthropic places best on 13 , Google places best on 7 and Moonshot AI places best on 1 . All 3 ranked on 19 of 28 boards.
Board Read Moonshot AI Anthropic Google 2 Text Arena Arena rating 1d yesterday Nov 9 1d yesterday #16 1,488 Kimi K3 #4 1,505 Claude Opus 5.5 #1 1,526 best of the picked Gemini 4 Carbon MultiChallenge Score 1d yesterday Nov 9 1d yesterday #15 61.4% Kimi K2.5 #10 67.7% Claude Haiku 5.5 #2 78.7% best of the picked Gemini 4 Argon 7 AA Intelligence Index 1d yesterday Nov 9 1d yesterday #10 46.5 Kimi K3.1 Pro #2 59.1 Claude Opus 5.5 #1 60.0 best of the picked Gemini 4 Carbon GPQA Diamond Accuracy 1d yesterday Nov 9 1d yesterday #11 94.2% Kimi K3.1 Pro Together #3 95.6% Claude Sonnet 5.5 #1 96.4% best of the picked Gemini 4 Argon LiveBench Global average 1d yesterday Nov 9 1d yesterday #14 79.2% Kimi K3 #1 83.4% best of the picked Claude Fable 5.1 #15 78.8% Gemini 3.7 Flash MMLU-Pro Accuracy · stale 1d yesterday Nov 9 1d yesterday #56 81.0% Kimi K2 Thinking #4 89.1% Claude Opus 4.6 #1 91.2% best of the picked Gemini 3.1 Pro Preview HLE Accuracy 1d yesterday Nov 9 1d yesterday #17 24.4% Kimi K2.5 #2 46.5% best of the picked Claude Fable 5.1 #3 46.4% Gemini 3.1 Pro Preview Epoch ECI ECI 1d yesterday Nov 9 1d yesterday #21 158.5 Kimi K3 #3 168.4 best of the picked Claude Opus 5.5 #6 166.4 Gemini 4 Argon ARC-AGI-2 Score 1d yesterday Nov 9 1d yesterday #35 60.4% Kimi K3 #3 93.3% best of the picked Claude Opus 5.5 #5 91.3% Gemini 4 Argon 4 WebDev Arena Arena rating 1d yesterday Nov 9 1d yesterday #9 1,684 Kimi K3.1 Pro #1 1,814 best of the picked Claude Opus 5.5 #4 1,767 Gemini 4 Carbon SWE-Bench Pro Resolved 1d yesterday Nov 9 1d yesterday #6 88.2% Kimi K3 #1 98.0% best of the picked Claude Opus 5 #3 92.2% Gemini 4 Argon SWE-bench Verified Resolved · stale 1d yesterday Nov 9 1d yesterday #10 70.8% Kimi K2.5 #1 76.8% best of the picked Claude Opus 4.5 #2 75.8% Gemini 3 Flash Aider Polyglot Pass rate · stale 1d yesterday Nov 9 1d yesterday #19 59.1% Kimi K2 Thinking #10 72.0% Claude Opus 4 #3 83.1% best of the picked Gemini 2.5 Pro (Jun 2025) 3 BFCL Overall accuracy · stale 1d yesterday Nov 9 1d yesterday #9 59.1% Moonshotai Kimi K2 #1 77.5% best of the picked Claude Opus 4.5 #3 72.5% Gemini 3 Pro MCP Atlas Pass rate 1d yesterday Nov 9 1d yesterday #11 82.3% Kimi K3 #3 87.2% Claude Fable 5.1 #1 90.7% best of the picked Gemini 4 Argon Search Arena Arena rating 1d yesterday Nov 9 1d yesterday #5 1,229 best of the picked Claude Fable 5 #14 1,208 Gemini 3.1 Pro Preview 2 Vision Arena Arena rating 1d yesterday Nov 9 1d yesterday #33 1,264 Kimi K2.6 #1 1,309 best of the picked Claude Fable 5 #6 1,294 Gemini 3.7 Flash Document Arena Arena rating 1d yesterday Nov 9 1d yesterday #29 1,451 Kimi K2.6 #1 1,499 best of the picked Claude Fable 5 #16 1,469 Gemini 3.8 Flash 2 Text-to-Image Arena rating 1d yesterday Nov 9 1d yesterday #7 1,325 Gemini Nano Banana 2.1 Image Edit Arena rating 1d yesterday Nov 9 1d yesterday #8 1,427 Gemini Nano Banana 2.1 2 Text-to-Video Arena rating 1d yesterday Nov 9 1d yesterday #1 1,541 Gemini Omni 1.2 Flash Image-to-Video Arena rating 1d yesterday Nov 9 1d yesterday #1 1,502 Gemini Omni 1.2 Flash 2 Speech Arena Arena rating 1d yesterday Nov 9 1d yesterday #6 1,283 Gemini 3.8 Flash TTS Open ASR Average WER 1d yesterday Nov 9 1d yesterday #58 13.20 WER gemma-4-E4B-it (google) 1 MTEB Mean task score 1d yesterday Nov 9 1d yesterday #5 68.4% gemini-embedding-001 (google) 1 AA Speed Output speed 1d yesterday Nov 9 1d yesterday #6 36 tok/s best of the picked Kimi K3.1 Pro #9 31 tok/s Claude Sonnet 5.5 #7 35 tok/s Gemini 4 Carbon 1 AA Price Blended price 1d yesterday Nov 9 1d yesterday #10 $4.00 Kimi K3.1 Pro #1 $0.20 best of the picked Claude Haiku 5.5 #5 $1.50 Gemini 3.8 Flash 1 OpenRouter Weekly tokens 1d yesterday Nov 9 1d yesterday #8 4.4T tok Kimi K3.1 Pro #6 6.8T tok best of the picked Claude Opus 5.5
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.
Modelsby Glyph