Leaderboards

Benchmarks

Every score carries its origin.

Reasoning374Coding337Math221Agentic375Chat636Speech778
Arena Code (WebDev)95Arena Coding168LiveBench Coding58LiveCodeBench16

Arena Code (WebDev)

Arena Score from human votes on web-app building tasks — not a percentage
Provenance95 modelsscores Aug 10, 2026 – Sep 25, 2026
1Claude Opus 5.5AnthropicScore1827Score / $91ScoredSep 25, 20262GPT-6 AstraOpenAIScore1792Score / $72ScoredSep 25, 20263Claude Fable 5.1AnthropicScore1751Score / $35ScoredSep 25, 20264Claude Opus 5AnthropicScore1693Score / $68ScoredSep 25, 20265GPT-6 SolOpenAIScore1681Score / $336ScoredSep 25, 20266Qwen3.8 MaxQwenScore1672Score / $338ScoredSep 25, 20267Kimi K3Moonshot AI2.8TScore1660Score / $169Score / GB VRAM~0.95ScoredSep 25, 20268Muse Spark 1.3MetaScore1656Score / $390ScoredSep 25, 20269Qwen3.8 FlashQwenScore1636Score / $4282ScoredSep 25, 202610Hy4 previewTencent780BScore1631Score / $724Score / GB VRAM~3.3ScoredSep 25, 202611Grok 4.7xAIScore1629Score / $272ScoredSep 25, 202612Claude Fable 5AnthropicScore1628Score / $33ScoredSep 11, 202613DeepSeek V4.1 FlashDeepSeek763BScore1621Score / $4052Score / GB VRAM~3.4ScoredSep 25, 202614Grok 4.6xAIScore1621Score / $270ScoredSep 25, 202615GLM 5.3Z.aiScore1619Score / $1420ScoredSep 25, 202616GPT-5.6 SolOpenAIScore1617Score / $323ScoredSep 25, 202617GLM 5.3 FlashZ.ai321BScore1613Score / $6453Score / GB VRAM~8ScoredSep 25, 202618GLM 5.2Z.ai753BScore1602Score / $1761Score / GB VRAM~3.4ScoredSep 25, 202619Gemini 3.7 FlashGoogleScore1593Score / $850ScoredSep 25, 202620GPT-6 LunaOpenAIScore1593Score / $6371ScoredSep 25, 202621Qwen3.8 27BQwen27.8BScore1591Score / $1068Score / GB VRAM~91ScoredSep 25, 202622DeepSeek V4 ProDeepSeek1.6TScore1582Score / $1010Score / GB VRAM~1.6ScoredSep 25, 202623DeepSeek V4 FlashDeepSeek291BScore1581Score / $18888Score / GB VRAM~8.6ScoredSep 25, 202624Gemini 3.8 FlashGoogleScore1580Score / $843ScoredSep 25, 202625Claude Opus 4.7AnthropicScore1556Score / $62ScoredSep 25, 202626Grok 4.5xAIScore1552Score / $259ScoredSep 25, 202627Muse Spark 1.1MetaScore1541Score / $363ScoredSep 25, 202628Claude Sonnet 5AnthropicScore1539Score / $154ScoredSep 25, 202629Claude Opus 4.6AnthropicScore1537Score / $61ScoredSep 25, 202630Gemini 3.6 FlashGoogleScore1536Score / $819ScoredAug 10, 202631Claude Opus 4.8AnthropicScore1535Score / $61ScoredSep 25, 202632Muse Spark 1.2MetaScore1531Score / $360ScoredSep 25, 202633Claude Sonnet 4.6AnthropicScore1521Score / $101ScoredSep 25, 202634GPT-5.6 TerraOpenAIScore1521Score / $253ScoredSep 25, 202635GPT-5.6 LunaOpenAIScore1520Score / $2533ScoredSep 25, 202636Qwen3.7 MaxQwenScore1515Score / $404ScoredSep 25, 202637GPT-5.5OpenAIScore1512Score / $101ScoredSep 25, 202638Hy3Tencent299BScore1509Score / $2858Score / GB VRAM~8ScoredSep 25, 202639Kimi K2.6Moonshot AI1.1TScore1509Score / $825Score / GB VRAM~2.3ScoredSep 25, 202640GLM 5.1Z.ai754BScore1508Score / $497Score / GB VRAM~3.2ScoredSep 25, 202641Gemini 3.5 FlashGoogleScore1499Score / $333ScoredSep 25, 202642MiniMax M3MiniMax427BScore1483Score / $1545Score / GB VRAM~5.5ScoredSep 25, 202643Qwen3.6 Max PreviewQwenScore1481Score / $240ScoredSep 25, 202644MiMo-V2.5-ProXiaomi1TScore1477Score / $2426Score / GB VRAM~2.3ScoredSep 25, 202645Kimi K2.7 CodeMoonshot AI1.1TScore1473Score / $491Score / GB VRAM~2.2ScoredSep 25, 202646Claude Opus 4.5AnthropicScore1468Score / $59ScoredSep 25, 202647Qwen3.6 PlusQwenScore1461Score / $749ScoredSep 25, 202648Gemini 3.1 Pro Preview Custom ToolsGoogleScore1446Score / $241ScoredSep 25, 202649Gemini 3.5 Flash LiteGoogleScore1439Score / $1151ScoredSep 25, 202650MiMo-V2.5Xiaomi311BScore1438Score / $6040Score / GB VRAM~7.3ScoredSep 25, 202651Kimi K2 ThinkingMoonshot AI1.1TScore1436Score / $574Score / GB VRAM~2.2ScoredSep 23, 202652GLM 4.7Z.ai358BScore1435Score / $820Score / GB VRAM~6.4ScoredSep 25, 202653GLM 5Z.ai754BScore1434Score / $747Score / GB VRAM~3ScoredSep 25, 202654GPT-5OpenAIScore1418Score / $142ScoredSep 25, 202655GPT-5.2 ChatOpenAIScore1415Score / $202ScoredSep 25, 202656InklingThinking Machines952BScore1411Score / $348Score / GB VRAM~2.3ScoredSep 25, 202657GPT-5.3-CodexOpenAIScore1408Score / $101ScoredSep 25, 202658Inkling SmallThinking Machines266BScore1408Score / $1173Score / GB VRAM~8.4ScoredSep 25, 202659Kimi K2.5Moonshot AI1.1TScore1404Score / $624Score / GB VRAM~2.1ScoredSep 25, 202660GLM 5V TurboZ.aiScore1400Score / $350ScoredSep 25, 202661Qwen3.5 397B A17BQwen403BScore1400Score / $598Score / GB VRAM~5.5ScoredSep 25, 202662GPT-5.4 MiniOpenAIScore1397Score / $621ScoredSep 25, 202663MiniMax M2.7MiniMax229BScore1397Score / $1663Score / GB VRAM~9.7ScoredSep 25, 202664GPT-5.4OpenAIScore1394Score / $186ScoredSep 25, 202665Claude Opus 4.1AnthropicScore1390Score / $19ScoredSep 25, 202666MiniMax M2.5MiniMax229BScore1386Score / $1459Score / GB VRAM~9.6ScoredSep 25, 202667Claude Sonnet 4.5AnthropicScore1384Score / $92ScoredSep 25, 202668Gemma 4 31BGoogle31.3BScore1363Score / $4132Score / GB VRAM~69ScoredSep 25, 202669Qwen3.5-122B-A10BQwen125BScore1360Score / $654Score / GB VRAM~17ScoredSep 25, 202670Gemma 4 26B A4BGoogle25.8BScore1358Score / $6175Score / GB VRAM~83ScoredSep 25, 202671Qwen3.5-27BQwen27.8BScore1357Score / $870Score / GB VRAM~77ScoredSep 25, 202672Grok 4.3xAIScore1357Score / $543ScoredSep 25, 202673Muse Glimmer 30BMeta29.8BScore1355Score / $1232Score / GB VRAM~72ScoredSep 25, 202674Hy3 previewTencent299BScore1355Score / $2259Score / GB VRAM~7.2ScoredSep 25, 202675Laguna M.1poolside226BScore1348Score / $3371Score / GB VRAM~9.5ScoredSep 25, 202676GPT-5.1OpenAIScore1341Score / $268ScoredSep 25, 202677GLM 4.6Z.ai357BScore1339Score / $765Score / GB VRAM~6ScoredSep 25, 202678GPT-5.2-CodexOpenAIScore1339Score / $96ScoredSep 25, 202679GPT-5.1-CodexOpenAIScore1337Score / $134ScoredSep 25, 202680Claude Haiku 4.5AnthropicScore1330Score / $266ScoredSep 25, 202681DeepSeek V3.2DeepSeek685BScore1325Score / $4280Score / GB VRAM~3.1ScoredSep 25, 202682MiniMax M2MiniMax229BScore1297Score / $1271Score / GB VRAM~9ScoredSep 25, 202683Qwen3 Coder 480B A35BQwen480BScore1274Score / $1274Score / GB VRAM~4.2ScoredSep 25, 202684DeepSeek V3.2 ExpDeepSeek685BScore1272Score / $3102Score / GB VRAM~2.9ScoredSep 25, 202685Mistral Medium 3.5Mistral AI128BScore1263Score / $168Score / GB VRAM~16ScoredSep 25, 202686Gemini 3.1 Flash Lite PreviewGoogleScore1256Score / $1675ScoredSep 25, 202687Qwen3.5-35B-A3BQwen36BScore1254Score / $1671Score / GB VRAM~55ScoredSep 25, 202688GPT-5.1-Codex-MiniOpenAIScore1244Score / $622ScoredSep 25, 202689Qwen3.5-FlashQwenScore1243Score / $4782ScoredSep 25, 202690Trinity Large ThinkingArcee AI399BScore1238Score / $1547Score / GB VRAM~4.9ScoredSep 25, 202691Mistral Large 3 2512Mistral AIScore1230Score / $820ScoredSep 25, 202692Gemini 2.5 ProGoogleScore1227Score / $245ScoredSep 25, 202693Devstral 2 2512Mistral AI125BScore1196Score / $598Score / GB VRAM~15ScoredSep 25, 202694Granite 4.1 8BIBM8.8BScore1191Score / $11909Score / GB VRAM~210ScoredSep 25, 202695Mercury 2InceptionScore1170Score / $1560ScoredSep 25, 2026