LMArena WebDev Elo Leaderboard
coding115 models · Updated October 2026
On the LMArena WebDev Elo benchmark, Claude Opus 5.5 ranks #1 with a score of 1811, while Granite 4.1 8B offers the best score-per-dollar at $0.10/1M output tokens. The full ranking, with cost per million tokens, is below.
💰 Best value
Granite 4.1 8B — score 1191 at $0.10/1M output tokens
| # | Model | Score | Output / 1M | Context |
|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 1811 | $20.00 | 1M |
| 2 | GPT-6 AstraOpenAI | 1780 | $50.00 | 1.1M |
| 3 | Claude Sonnet 5.5Anthropic | 1774 | $10.00 | 1M |
| 4 | GPT-6.1 SolOpenAI | 1760 | $10.00 | 1.1M |
| 5 | Claude Fable 5.1Anthropic | 1743 | $50.00 | 1M |
| 6 | Claude Opus 5Anthropic | 1690 | $25.00 | 1M |
| 7 | GPT-6 SolOpenAI | 1686 | $10.00 | 1.1M |
| 8 | Gemini 4 ArgonGoogle | 1678 | $10.00 | 1M |
| 9 | Qwen3.8 MaxAlibaba | 1673 | $6.00 | 1M |
| 10 | Muse Spark 1.3Meta | 1659 | $4.25 | 1M |
| 11 | Kimi K3MoonshotAI | 1655 | $15.00 | 1M |
| 12 | Muse WonderMeta | 1651 | $20.00 | 1M |
| 13 | Grok 4.7SpaceXAI | 1640 | $6.00 | 500K |
| 14 | MiMo-V2.6-FlashXiaomi | 1635 | $0.28 | 1.1M |
| 15 | Qwen3.8 Flash NextAMD | 1633 | $0.47 | 262K |
| 16 | Hy4 previewTencent | 1632 | $2.50 | 1M |
| 17 | MiMo-V2.6-ProXiaomi | 1625 | $0.87 | 1.1M |
| 18 | Claude Fable 5Anthropic | 1624 | $50.00 | 1M |
| 19 | GLM 5.3Z.ai | 1621 | $4.80 | 1M |
| 20 | DeepSeek V4.1 FlashDeepSeek | 1619 | $0.89 | 1M |
| 21 | Grok 4.6SpaceXAI | 1617 | $6.00 | 500K |
| 22 | GPT-5.6 SolOpenAI | 1617 | $20.00 | 1.1M |
| 23 | GLM 5.3 FlashZ.ai | 1606 | $0.90 | 1M |
| 24 | GLM 5.2Z.ai | 1604 | $7.00 | 1M |
| 25 | Qwen3.8 27BQwen | 1594 | $2.20 | 1M |
| 26 | Gemini 3.7 FlashGoogle | 1590 | $3.75 | 1M |
| 27 | Claude Haiku 5.5Anthropic | 1585 | $0.50 | 1M |
| 28 | Gemini 3.8 FlashGoogle | 1585 | $3.75 | 1M |
| 29 | DeepSeek V4 Pro 0423DeepSeek | 1582 | $1.91 | 1M |
| 30 | GPT-6 LunaOpenAI | 1582 | $0.50 | 1.1M |
| 31 | DeepSeek V4 Flash 0423DeepSeek | 1582 | $0.18 | 1M |
| 32 | Step 5 PreviewStepFun | 1564 | $2.70 | 1M |
| 33 | Claude Opus 4.7Anthropic | 1558 | $25.00 | 1M |
| 34 | Claude Opus 4.8Anthropic | 1556 | $25.00 | 1M |
| 35 | Grok 4.5SpaceXAI | 1553 | $6.00 | 500K |
| 36 | Claude Opus 4.6Anthropic | 1546 | $25.00 | 1M |
| 37 | Mistral Large 4Mistral | 1542 | $2.09 | 1M |
| 38 | Muse Spark 1.1Meta | 1542 | $4.25 | 1M |
| 39 | Claude Sonnet 5Anthropic | 1540 | $10.00 | 1M |
| 40 | Gemini 3.6 FlashGoogle | 1538 | $3.75 | 1M |
| 41 | Muse Spark 1.2Meta | 1533 | $4.25 | 1M |
| 42 | Claude Sonnet 4.6Anthropic | 1522 | $15.00 | 1M |
| 43 | GPT-5.6 TerraOpenAI | 1522 | $12.00 | 1.1M |
| 44 | GPT-5.6 LunaOpenAI | 1518 | $1.20 | 1.1M |
| 45 | Qwen3.7 MaxQwen | 1515 | $4.43 | 1M |
| 46 | GPT-5.5OpenAI | 1513 | $30.00 | 1.1M |
| 47 | Kimi K2.6MoonshotAI | 1509 | $3.41 | 262K |
| 48 | GLM 5.1Z.ai | 1508 | $3.04 | 205K |
| 49 | Hy3Tencent | 1507 | $0.53 | 262K |
| 50 | Gemini 3.5 FlashGoogle | 1499 | $9.00 | 1M |
| 51 | Claude Opus 4.5Anthropic | 1493 | $25.00 | 200K |
| 52 | MiniMax M3MiniMax | 1482 | $1.20 | 1M |
| 53 | Qwen3.6 Max PreviewAlibaba | 1482 | $7.80 | 262K |
| 54 | MiMo-V2.5-ProXiaomi | 1480 | $0.87 | 1.1M |
| 55 | Kimi K2.7 CodeMoonshotAI | 1473 | $3.35 | 262K |
| 56 | GPT-5.4OpenAI | 1463 | $15.00 | 1.1M |
| 57 | Qwen3.6 PlusQwen | 1461 | $1.95 | 1M |
| 58 | Gemini 3.1 Pro PreviewGoogle | 1447 | $12.00 | 1M |
| 59 | Gemini 3.5 Flash LiteGoogle | 1440 | $2.50 | 1M |
| 60 | Gemini 3 FlashOpenCode | 1440 | $3.00 | 1M |
| 61 | Gemini 3 ProOpenCode | 1440 | $12.00 | 1M |
| 62 | MiMo-V2.5Xiaomi | 1438 | $0.28 | 1.1M |
| 63 | Kimi K2.5MoonshotAI | 1436 | $3.00 | 262K |
| 64 | GLM 5Z.ai | 1435 | $3.20 | 205K |
| 65 | GLM 4.7Z.ai | 1435 | $2.20 | 205K |
| 66 | MiMo-V2-ProXiaomi | 1434 | $0.87 | 1M |
| 67 | GPT-5OpenAI | 1417 | $10.00 | 400K |
| 68 | GPT-5.2OpenAI | 1417 | $14.00 | 400K |
| 69 | InklingThinking Machines | 1413 | $4.05 | 524K |
| 70 | Inkling SmallThinking Machines | 1408 | $1.20 | 524K |
| 71 | GPT-5.3-CodexOpenAI | 1406 | $14.00 | 400K |
| 72 | GLM 5V TurboZ.ai | 1401 | $4.00 | 203K |
| 73 | Qwen3.5 397B-A17BAlibaba | 1400 | $3.60 | 262K |
| 74 | MiniMax M2.7MiniMax | 1398 | $1.20 | 205K |
| 75 | GPT-5.4 MiniOpenAI | 1398 | $4.50 | 400K |
| 76 | GPT-5.1OpenAI | 1395 | $10.00 | 400K |
| 77 | Claude Sonnet 4.5Anthropic | 1393 | $15.00 | 1M |
| 78 | Claude Opus 4.1Anthropic | 1390 | $75.00 | 200K |
| 79 | MiniMax M2.5MiniMax | 1386 | $0.95 | 205K |
| 80 | grok-4.20-beta-0309-reasoning302.AI | 1375 | $6.00 | 2M |
| 81 | Solar Pro 4Upstage | 1371 | $0.36 | 524K |
| 82 | Gemma 4 31BGoogle | 1366 | $0.40 | 262K |
| 83 | DeepSeek V3.2DeepSeek | 1362 | $0.80 | 164K |
| 84 | Qwen3.5 122B-A10BAlibaba | 1361 | $3.20 | 262K |
| 85 | Gemma 4 26B A4BGoogle | 1359 | $0.26 | 262K |
| 86 | Qwen3.5 27BAlibaba | 1358 | $2.40 | 262K |
| 87 | Grok 4.3SpaceXAI | 1357 | $2.50 | 1M |
| 88 | Laguna M.1Poolside | 1349 | $0.40 | 262K |
| 89 | GLM 4.6Z.ai | 1341 | $2.00 | 205K |
| 90 | GPT-5.2-CodexOpenAI | 1338 | $14.00 | 400K |
| 91 | GPT-5.1-CodexOpenAI | 1337 | $10.00 | 400K |
| 92 | MiMo-V2-FlashXiaomi | 1330 | $0.28 | 262K |
| 93 | Claude Haiku 4.5Anthropic | 1330 | $5.00 | 200K |
| 94 | Kimi K2 Thinking TurboMoonshot AI | 1322 | $8.00 | 262K |
| 95 | Laguna XS.2Poolside | 1303 | $0.20 | 262K |
| 96 | MiniMax M2MiniMax | 1297 | $1.02 | 205K |
| 97 | Qwen3-Coder 480B-A35B InstructAlibaba | 1275 | $7.50 | 262K |
| 98 | DeepSeek V3.2 ExpDeepSeek | 1272 | $0.41 | 164K |
| 99 | Mistral Medium 3.5Mistral | 1263 | $7.50 | 262K |
| 100 | Gemini 3.1 Flash Lite PreviewGoogle | 1256 | $1.50 | 1M |
| 101 | KAT-Coder-Pro V1Vercel AI Gateway | 1256 | $1.20 | 256K |
| 102 | Qwen3.5 35B-A3BAlibaba | 1254 | $2.00 | 262K |
| 103 | Qwen3.5 FlashAlibaba | 1243 | $0.40 | 1M |
| 104 | GPT-5.1-Codex-MiniOpenAI | 1243 | $2.00 | 400K |
| 105 | Grok 4.1 FastOfox | 1242 | $0.50 | 2M |
| 106 | Trinity Large ThinkingArcee AI | 1238 | $0.80 | 262K |
| 107 | Mistral Large 3 2512Mistral | 1230 | $1.50 | 262K |
| 108 | Gemini 2.5 ProGoogle | 1226 | $10.00 | 1M |
| 109 | grok-4.1302.AI | 1214 | $10.00 | 200K |
| 110 | Devstral 2 2512Mistral | 1196 | $2.00 | 262K |
| 111 | Granite 4.1 8BIBM | 1191 | $0.10 | 131K |
| 112 | Mercury 2Inception | 1171 | $0.75 | 128K |
| 113 | Grok Code Fast 1OpenCode | 1166 | $0.00 | 256K |
| 114 | Grok 4 FastZenMux | 1159 | $0.50 | 2M |
| 115 | Devstral MediumMistral | 1071 | $2.00 | 128K |
What does LMArena WebDev Elo test?
Overview This dataset contains ALL in-the-wild conversation crowdsourced from Search Arena between March 18, 2025 and May 8, 2025. It includes 24,069 multi-turn conversations with search-LLMs across diverse intents, languages, and topics—alongs...
Frequently asked questions
What is the LMArena WebDev Elo benchmark?
Overview This dataset contains ALL in-the-wild conversation crowdsourced from Search Arena between March 18, 2025 and May 8, 2025. It includes 24,069 multi-turn conversations with search-LLMs across diverse intents, languages, and topics—alongs...
Which model is best on LMArena WebDev Elo?
Claude Opus 5.5 (Anthropic) currently leads the LMArena WebDev Elo leaderboard with a score of 1811.
What is the best value model on LMArena WebDev Elo?
Granite 4.1 8B offers the best score-per-dollar on LMArena WebDev Elo — a score of 1191 at $0.10 per 1M output tokens.
Pricing is indicative — confirm with the provider before production use. Updated October 2026.