AI Rankings
Best AI for Web Development: Ranked by People Who Judge the Results
On Arena's web development board, people describe a web page or small app, two anonymous models build it, and the voter picks the better result. This page ranks models by the rating those votes produce.
Data updated
Top 20 at a glance
Tap a model to see its full page: scores, price and how it compares.
Bars start at 1590, not zero, so differences are easier to see.
Full ranking
All 85 models.Download the full table (CSV)Data credit: Arena (formerly LMArena), Epoch AI, LiteLLM model price list, used under open licences (details at the foot of the page).
| Rank and model | Arena web development rating | Likely range | Votes | Epoch index | Price per 1M tokens (in / out) |
|---|---|---|---|---|---|
1Claude Opus 5.5Anthropic | 1807 | 1794 to 1820 | 3,140 | 168.8 | $4.00 / $20.00 |
2GPT-6 AstraOpenAI | 1774 | 1764 to 1784 | 6,579 | 168.8 | $10.00 / $50.00 |
3Claude Sonnet 5.5Anthropic | 1769 | 1757 to 1781 | 3,822 | 166.2 | $2.00 / $10.00 |
4GPT-6.1 SolOpenAI | 1749 | 1737 to 1762 | 3,469 | 168.4 | $2.00 / $10.00 |
5Claude Fable 5.1Anthropic | 1743 | 1734 to 1751 | 8,098 | 166.5 | $10.00 / $50.00 |
6Claude Opus 5Anthropic | 1691 | 1685 to 1697 | 18,995 | 163.9 | $5.00 / $25.00 |
7GPT-6 SolOpenAI | 1681 | 1671 to 1691 | 5,008 | 163.7 | $2.00 / $10.00 |
8Gemini 4 ArgonGoogle | 1674 | 1662 to 1687 | 2,729 | 166.4 | $2.00 / $10.00 |
9Qwen3.8 Max (0902)Alibaba | 1673 | 1666 to 1680 | 15,637 | 155.1 | Not measured |
10Qwen 3.8 MaxAlibaba | 1672 | 1660 to 1684 | 3,429 | 157.4 | $2.00 / $6.00 |
11Muse Spark 1.3Meta | 1657 | 1650 to 1665 | 9,921 | 157.2 | $1.25 / $4.25 |
12Kimi K3Moonshot AI · Open weights | 1655 | 1649 to 1661 | 20,312 | 157.9 | $3.00 / $15.00 |
13Muse WonderMeta | 1651 | 1638 to 1664 | 2,733 | 161.8 | Not measured |
14Grok 4.7xAI | 1638 | 1629 to 1647 | 5,725 | 154.4 | $2.00 / $6.00 |
15MiMo-V2.6-ProXiaomi · Open weights | 1625 | 1616 to 1635 | 5,678 | Not measured | $0.43 / $0.87 |
16Claude Fable 5Anthropic | 1624 | 1618 to 1630 | 16,342 | 163.4 | $10.00 / $50.00 |
17GLM-5.3Z.ai · Open weights | 1623 | 1615 to 1630 | 10,315 | 156.0 | $1.40 / $4.40 |
18DeepSeek V4.1 FlashDeepSeek · Open weights | 1619 | 1608 to 1630 | 3,530 | 155.9 | $0.30 / $1.20 |
19Grok 4.6xAI | 1617 | 1610 to 1624 | 11,022 | 156.6 | $2.00 / $6.00 |
20GPT-5.6 SolOpenAI | 1615 | 1609 to 1621 | 18,243 | 162.8 | $4.00 / $20.00 |
21GLM-5.3-FlashZ.ai · Open weights | 1608 | 1601 to 1614 | 19,680 | 153.1 | $0.15 / $0.50 |
22GLM-5.2Z.ai · Open weights | 1604 | 1598 to 1610 | 17,082 | 151.9 | $1.40 / $4.40 |
23Qwen 3.8 27BAlibaba · Open weights | 1591 | 1585 to 1598 | 17,186 | 151.7 | $0.45 / $3.00 |
24Gemini 3.7 FlashGoogle | 1590 | 1584 to 1597 | 13,307 | 158.0 | $0.75 / $3.75 |
25Claude Haiku 5.5Anthropic | 1586 | 1575 to 1596 | 3,917 | 159.3 | $0.10 / $0.50 |
26Gemini 3.8 FlashGoogle | 1585 | 1578 to 1592 | 11,405 | 157.4 | $0.75 / $3.75 |
27DeepSeek V4 Pro 0813DeepSeek · Open weights | 1582 | 1573 to 1592 | 4,714 | 155.6 | $1.32 / $3.96 |
28DeepSeek-V4-FlashDeepSeek · Open weights | 1582 | 1573 to 1591 | 4,939 | 146.1 | $0.30 / $1.20 |
29GPT-6 LunaOpenAI | 1579 | 1573 to 1585 | 24,354 | 156.8 | $0.10 / $0.50 |
30Claude Opus 4.7Anthropic | 1558 | 1552 to 1564 | 17,920 | 156.8 | $5.00 / $25.00 |
31Claude Opus 4.8Anthropic | 1556 | 1550 to 1562 | 18,486 | 159.0 | $5.00 / $25.00 |
32Grok 4.5xAI | 1553 | 1546 to 1560 | 11,083 | 154.1 | $2.00 / $6.00 |
33Claude Opus 4.6Anthropic | 1546 | 1541 to 1552 | 19,404 | 155.7 | $5.00 / $25.00 |
34Mistral Large 4Mistral AI | 1545 | 1533 to 1556 | 3,269 | 150.0 | $0.68 / $2.09 |
35Muse Spark 1.1Meta | 1543 | 1535 to 1550 | 7,970 | 154.1 | $1.25 / $4.25 |
36Claude Sonnet 5Anthropic | 1541 | 1535 to 1546 | 16,187 | 156.9 | $2.00 / $10.00 |
37Gemini 3.6 FlashGoogle | 1538 | 1531 to 1546 | 8,758 | 154.5 | $0.75 / $3.75 |
38Muse Spark 1.2Meta | 1533 | 1520 to 1547 | 2,163 | 155.1 | $1.25 / $4.25 |
39Muse Spark OpenMeta · Open weights | 1526 | 1510 to 1543 | 1,476 | 155.1 | $1.25 / $4.25 |
40Claude Sonnet 4.6Anthropic | 1523 | 1518 to 1528 | 22,852 | 152.8 | $3.00 / $15.00 |
41GPT-5.6 TerraOpenAI | 1521 | 1514 to 1527 | 14,450 | 160.1 | $2.00 / $12.00 |
42GPT-5.6 LunaOpenAI | 1519 | 1513 to 1525 | 12,575 | 156.5 | $0.20 / $1.20 |
43Qwen3.7-MaxAlibaba | 1515 | 1508 to 1523 | 8,888 | 154.0 | $2.50 / $7.50 |
44GPT-5.5OpenAI | 1513 | 1507 to 1519 | 15,620 | 159.7 | $5.00 / $30.00 |
45Kimi K2.6Moonshot AI · Open weights | 1509 | 1502 to 1517 | 10,172 | 150.9 | $0.95 / $4.00 |
46GLM-5.1Z.ai · Open weights | 1508 | 1501 to 1515 | 10,703 | 149.8 | $1.40 / $4.40 |
47Gemini 3.5 FlashGoogle | 1499 | 1492 to 1506 | 9,382 | 154.9 | $1.50 / $9.00 |
48Claude Opus 4.5Anthropic | 1494 | 1486 to 1502 | 11,033 | 150.1 | $5.00 / $25.00 |
49MiniMax-M3MiniMax · Open weights | 1482 | 1477 to 1488 | 20,251 | 146.9 | $0.30 / $1.20 |
50Qwen 3.6 Max (Preview)Alibaba | 1482 | 1470 to 1495 | 2,716 | 149.4 | $1.27 / $7.61 |
51MiMo-V2.5-ProXiaomi · Open weights | 1481 | 1475 to 1486 | 22,766 | Not measured | $0.52 / $1.04 |
52Kimi K2.7 CodeMoonshot AI · Open weights | 1474 | 1464 to 1483 | 4,961 | 150.1 | $0.95 / $4.00 |
53DeepSeek-V4-ProDeepSeek · Open weights | 1464 | 1458 to 1471 | 13,164 | 149.1 | $1.32 / $3.96 |
54GPT-5.4OpenAI | 1464 | 1445 to 1483 | 1,351 | 157.3 | $2.50 / $15.00 |
55Qwen 3.6 PlusAlibaba | 1462 | 1456 to 1467 | 18,585 | 147.3 | $0.50 / $3.00 |
56Gemini 3.1 ProGoogle | 1447 | 1442 to 1452 | 24,232 | 155.3 | $2.00 / $12.00 |
57Gemini 3 ProGoogle | 1440 | 1432 to 1449 | 14,098 | 153.1 | $2.00 / $12.00 |
58Gemini 3 FlashGoogle | 1440 | 1432 to 1449 | 10,893 | 151.9 | $0.50 / $3.00 |
59Kimi K2.5Moonshot AI · Open weights | 1437 | 1431 to 1442 | 20,102 | 148.1 | $0.60 / $3.00 |
60GLM-4.7Z.ai · Open weights | 1435 | 1423 to 1448 | 4,007 | 143.6 | $0.60 / $2.20 |
60GLM-5Z.ai · Open weights | 1435 | 1427 to 1444 | 7,497 | 145.8 | $1.00 / $3.20 |
62Gemini 3.5 Flash-LiteGoogle | 1434 | 1393 to 1475 | 237 | 144.9 | $0.30 / $2.50 |
63GPT-5OpenAI | 1418 | 1402 to 1433 | 3,092 | 150.0 | $1.25 / $10.00 |
64GPT-5.2OpenAI | 1417 | 1394 to 1439 | 1,082 | 153.7 | $1.75 / $14.00 |
65InklingThinking Machines · Open weights | 1415 | 1409 to 1421 | 16,256 | 148.4 | $1.00 / $4.05 |
66Inkling-SmallThinking Machines · Open weights | 1409 | 1400 to 1418 | 5,329 | 150.3 | $0.47 / $1.20 |
67GPT-5.3 CodexOpenAI | 1407 | 1393 to 1420 | 2,641 | 157.1 | $1.75 / $14.00 |
68Qwen3.5 397B-A17BAlibaba · Open weights | 1400 | 1395 to 1405 | 21,176 | 146.8 | $0.60 / $3.60 |
69MiniMax-M2.7MiniMax · Open weights | 1398 | 1392 to 1404 | 15,975 | 145.8 | $0.30 / $1.19 |
70GPT-5.4 MiniOpenAI | 1398 | 1391 to 1404 | 11,959 | 148.7 | $0.75 / $4.50 |
71GPT-5.1OpenAI | 1396 | 1384 to 1407 | 4,914 | 149.8 | $1.25 / $10.00 |
72Claude Sonnet 4.5Anthropic | 1394 | 1386 to 1402 | 13,255 | 146.7 | $3.00 / $15.00 |
73Claude Opus 4.1Anthropic | 1391 | 1380 to 1401 | 6,929 | 144.1 | $15.00 / $75.00 |
74MiniMax-M2.5MiniMax · Open weights | 1386 | 1378 to 1395 | 7,634 | 146.6 | $0.30 / $1.20 |
| 1376 | 1369 to 1382 | 15,469 | Not measured | $1.25 / $2.50 | |
76DeepSeek-V3.2DeepSeek · Open weights | 1363 | 1353 to 1372 | 6,925 | 146.1 | $0.28 / $0.40 |
77Gemma 4 26B A4BGoogle · Open weights | 1360 | 1343 to 1377 | 1,503 | 141.9 | $0.15 / $0.60 |
78Claude Haiku 4.5Anthropic | 1331 | 1326 to 1335 | 29,087 | 142.2 | $1.00 / $5.00 |
79DeepSeek-V3.2-ExpDeepSeek · Open weights | 1272 | 1258 to 1285 | 3,970 | 144.9 | $0.27 / $0.41 |
80Mistral Medium 3.5Mistral AI · Open weights | 1264 | 1249 to 1279 | 2,315 | 141.4 | $1.50 / $7.50 |
81Gemini 3.1 Flash-LiteGoogle | 1256 | 1250 to 1263 | 14,418 | 144.3 | $0.25 / $1.50 |
82Qwen3.5-35B-A3BAlibaba · Open weights | 1254 | 1237 to 1272 | 1,617 | 142.5 | $0.25 / $1.25 |
83Gemini 2.5 Pro (Jun 2025)Google | 1227 | 1211 to 1242 | 2,729 | 145.1 | $1.25 / $10.00 |
84Grok 4.1xAI | 1215 | 1190 to 1240 | 961 | Not measured | Not measured |
85Grok 4 FastxAI | 1160 | 1133 to 1187 | 743 | 144.1 | Not measured |
How to read this leaderboard
Higher is better, and only the gaps between models matter. Voters judge how the result looks and works, so this tells you about practical results rather than code style. Likely ranges that overlap mean the models are effectively tied. A model can rank well here and still be a poor fit for large existing codebases.
Questions people ask
What does the web development board measure?
It measures which model's web page or app people prefer when two anonymous models build from the same request. It rewards results that look right and work, as judged by the voter.
Is this the same as the coding leaderboard?
No. The coding board is about programming questions and answers. This board is about building something visible, such as a page or a small app. A model can rank differently on each.
Does the top model suit my work best?
Not always. A leaderboard compares models on one kind of test. Price, speed, privacy rules, the apps you already use and the type of work you do all matter. Use the rank as a starting point, then try two or three models on your own tasks.
Why is a model I know missing?
A model appears only when an open source has published results for it. New models can take days or weeks to be measured, and models with no published score are left out rather than guessed.
Where does the data come from?
Ratings come from the open leaderboard dataset published by Arena (arena.ai), licensed CC BY 4.0. We keep the latest published ratings, match models to our other sources and rank them. Arena does not endorse this site.
How often is this leaderboard updated?
We check our open data sources every day and show the date of the data at the top of the page. A model only moves when its source publishes new results, so a number can stay the same for weeks.
More leaderboards
Different questions need different tests. Pick the one that matches your work.
- Overall intelligenceWhich AI model is the smartest overall?Leads: Claude Opus 5.5 (168.8)270 models ranked
- Human preferenceWhich AI do people prefer in blind tests?Leads: Gemini 4 Argon (1517)166 models ranked
- CodingWhich AI model is best for coding?Leads: Gemini 4 Argon (1552)166 models ranked
- Image understandingWhich AI model understands images best?Leads: Claude Fable 5 (1309)85 models ranked
- Open-weights modelsWhich open-source AI model is best?Leads: Kimi K3 (157.9)123 models ranked
- Cheapest strong modelsWhat is the cheapest AI model that is still good?Cheapest: gpt-oss-20b ($0.10)119 models ranked
- Best valueWhich AI models give the most intelligence for the money?Line starts at: gpt-oss-20b ($0.10)7 models ranked
- Context windowWhich AI model can read the most text at once?Leads: GPT-5.4 (1.1M tokens)193 models ranked
- GPQA DiamondWhich AI model scores highest on GPQA Diamond?Leads: Gemini 4 Argon (96.4%)183 models ranked
- Humanity's Last ExamWhich AI model scores highest on Humanity's Last Exam?Leads: GPT-6 Astra (54.8%)41 models ranked
- SWE-bench VerifiedWhich AI model scores highest on SWE-bench Verified?Leads: Claude Opus 4.7 (83.5%)32 models ranked
- Aider PolyglotWhich AI model scores highest on Aider Polyglot?Leads: GPT-5 (88.0%)46 models ranked
- Terminal-BenchWhich AI model scores highest on Terminal-Bench?Leads: GPT-5.3 Codex (64.7%)25 models ranked
- FrontierMath (Tiers 1-3)Which AI model scores highest on FrontierMath (Tiers 1-3)?Leads: GPT-6 Astra (93.7%)87 models ranked
- ARC-AGI-2Which AI model scores highest on ARC-AGI-2?Leads: GPT-6 Astra (95.0%)86 models ranked
- SimpleQA VerifiedWhich AI model scores highest on SimpleQA Verified?Leads: Gemini 4 Argon (76.0%)82 models ranked
- SimpleBenchWhich AI model scores highest on SimpleBench?Leads: Claude Fable 5 (81.9%)80 models ranked
- DeepSWEWhich AI model scores highest on DeepSWE?Leads: GPT-6 Astra (74.1%)26 models ranked
Where this data comes from
- Arena (formerly LMArena): Human-preference ratings: Arena (arena.ai) leaderboard dataset, CC BY 4.0. We keep the latest published ratings, match them to models from other sources and rank them. Ratings are not changed. Arena does not endorse this site.
- Epoch AI: Benchmark data: Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks (CC BY 4.0). We match models across sources, rank them and show them in charts. Scores are not changed. Epoch AI does not endorse this site.
- LiteLLM model price list: Prices and context windows: LiteLLM model price list (github.com/BerriAI/litellm, MIT licence). We use the model maker's own price entry when there is one. Prices change often, so always confirm on the maker's site before you buy.
Newest data on this page: 26 Oct 2026. Prices are list prices and change often, so confirm on the maker's site before you buy. Sponsors never change a score or an order. How we calculate this · Report a mistake