Skip to content
IFAI

Model rankings

1447 models · 5 boards · 2 sources

WebDev arena · People compare web pages built by two models · Published August 30, 2026

Best web builder
claude-opus-5-maxAnthropic1,688Score
Best open model
kimi-k3-maxMoonshot AI1,674Score
Best from China
kimi-k3-maxMoonshot AI1,674Score
Most voted
claude-haiku-4-5-20251001Anthropic27,346Votes
Top three by Score
1claude-opus-5-maxAnthropic1,688
2kimi-k3-maxMoonshot AI1,674
3qwen3.8-maxAlibaba1,669
Top three open models
1kimi-k3-maxMoonshot AI · #2 overall1,674
2hy4-previewTencent · #5 overall1,631
3qwen3.8-flash-nextAlibaba · #8 overall1,622
This week's changes
hy4-previewCoding · #5New
GLM-5.3 (max)Intelligence · #6New
MAI-Image-2.5-ProImage · #7New
qwen3.8-flash-nextCoding · #8New
glm-5.3-flashCoding · #11New
RankChangeModelScore95% intervalVotesPrice $/M
1—claude-opus-5-maxClosedAnthropic1,688±810,098—
2—kimi-k3-maxOpenMoonshot AI1,674±114,546—
3—qwen3.8-maxClosedAlibaba1,669±123,219$3.0
4—claude-opus-5-highClosedAnthropic1,661±810,164—
5Newhy4-previewOpenTencent1,631±181,276—
6▼ 1grok-4.6-highClosedxAI1,629±171,532—
7▼ 1claude-fable-5ClosedAnthropic1,628±89,001—
8Newqwen3.8-flash-nextOpenAlibaba1,622±161,778—
9▼ 2gpt-5.6-sol-xhigh (codex-harness)ClosedOpenAI1,616±710,570—
10▼ 2glm-5.3-maxOpenZ.ai1,608±132,560—
11Newglm-5.3-flashOpenZ.ai1,606±161,593$0.24
12▼ 3qwen3.8-27bOpenAlibaba1,598±123,163$1.1
13▼ 3gemini-3.7-flash-highClosedGoogle1,587±123,009—
14▼ 3glm-5.2-maxOpenZ.ai1,585±89,494—
15▼ 3deepseek-v4-pro-high-20260813OpenDeepSeek · 1 more variant1,583±113,212—
16▼ 3deepseek-v4-flash-highOpenDeepSeek · 1 more variant1,580±113,852—
17▼ 3claude-opus-4-8-highClosedAnthropic1,563±712,537—
18▼ 3claude-opus-4-7ClosedAnthropic1,557±615,374—
19▼ 3claude-opus-4-7-highClosedAnthropic1,557±615,867—
20▼ 3grok-4.5ClosedxAI1,556±87,058—
21▼ 3claude-opus-4-6-highClosedAnthropic1,546±617,849—
22▼ 3claude-opus-4-8ClosedAnthropic1,540±711,495—
23▼ 3muse-spark-1.1ClosedMeta1,540±87,018—
24▼ 3gemini-3.6-flash-highClosedGoogle1,538±87,592—
25▼ 3claude-sonnet-5-highClosedAnthropic1,537±88,032—
26▼ 3claude-opus-4-6ClosedAnthropic1,536±619,016—
27▼ 3muse-spark-1.2 (xHigh)ClosedMeta1,534±142,069—
28▼ 2seed-2.1-pro-previewClosedByteDance1,522±88,331—
29▼ 4claude-sonnet-4-6ClosedAnthropic1,522±521,115—
30▼ 3gpt-5.6-terra-xhigh (codex-harness)ClosedOpenAI1,519±86,470—
31▼ 2gpt-5.6-luna-xhigh (codex-harness)ClosedOpenAI1,518±86,567—
32▼ 2qwen3.7-max-20260517ClosedAlibaba1,517±88,381—
33▼ 5hy3OpenTencent1,513±123,025—
34▼ 1gpt-5.5-xhigh (codex-harness)ClosedOpenAI1,510±614,450—
35▼ 3kimi-k2.6OpenMoonshot AI1,509±79,595—
36▼ 5glm-5.1OpenZ.ai1,509±79,828—
37▼ 3gemini-3.5-flash-highClosedGoogle1,501±88,359—
38▼ 3claude-opus-4-5-20251101-high-32kClosedAnthropic1,495±810,604—
39▼ 3gemini-3.5-flash-mediumClosedGoogle1,492±710,867—
40▼ 3minimax-m3OpenMiniMax1,487±711,481$0.52
41▼ 3gpt-5.5-high (codex-harness)ClosedOpenAI1,486±616,439—
42▼ 3qwen3.6-max-previewClosedAlibaba1,478±132,609—
43▼ 3mimo-v2.5-proOpenXiaomi1,476±614,728—
44▼ 3kimi-k2.7-codeOpenMoonshot AI1,473±104,684—
45▼ 3claude-opus-4-5-20251101ClosedAnthropic1,468±712,910—
46▼ 3gpt-5.4-high (codex-harness)ClosedOpenAI1,463±191,277—
47▼ 3qwen3.6-plusClosedAlibaba1,460±617,107—
48▼ 3gpt-5.5 (codex-harness)ClosedOpenAI1,458±614,506—
49▼ 3gemini-3.5-flash-liteClosedGoogle1,449±42228$0.85
50▼ 3gemini-3.1-pro-previewClosedGoogle1,446±522,390—

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.