Design scores

Human-preference votes on generated UI. Measures taste, not whether the code works.

65 results on 1 board · checked

How this category is sourced

Human-preference votes on generated UI. Measures taste, not whether the code works.

  • Every value is copied as the board publishes it, with its metric, harness, as-of date and a source link. Boards use different metrics, so numbers are never compared across boards.
  • No Ratelimit Now score: we do not normalise, weight or combine results.
  • Stale means the as-of date is more than 90 days old. Frozen means the board stopped scoring new models. Preliminary is the board's own label.
  • A model's score depends on the harness or agent it ran in. It is not a score for any tool on this site.
  • Open and Chinese models sit in the same tables, with the license the board lists. Hosting region is not confirmed for any model.

Category status in our data: covered (thin) · Per-category Design Arena boards (Website, UI Component, Mobile App, Android Native) load client-side and could not be fetched; only All-categories and Fullstack were captured.

Older versions = superseded by a newer model of the same family on the same board (the data's own flag).

Design Arena (All categories)

Elo (human preference)Harness: Design ArenaNewest as-of 2026-10-2419 latest-gen · 41 current · 24 older · 15 open weights

Human preferencedesignarena.ai
Top 10 on Design Arena (All categories)Bar height = Elo (human preference), axis starts at 1,300. Lab logo on top; small logo = harness tool. As of 2026-10-24.
1,3001,3461,392claude-opus-5-5: 1,389 (Elo (human preference)) 1,389claude-opus-5-5gpt-6-astra: 1,383 (Elo (human preference)) 1,383gpt-6-astramuse-wonder-1.0: 1,377 (Elo (human preference)) 1,377muse-wonder-1.0gpt-6-astra-max: 1,371 (Elo (human preference)) 1,371gpt-6-astra-maxkimi-k3: 1,359 (Elo (human preference)) CN 1,359kimi-k3gpt-6.1-sol: 1,358 (Elo (human preference)) 1,358gpt-6.1-solclaude-sonnet-5-5: 1,354 (Elo (human preference)) 1,354claude-sonnet-5-muse-spark-1.3: 1,350 (Elo (human preference)) 1,350muse-spark-1.3muse-spark-1.3-max: 1,349 (Elo (human preference)) 1,349muse-spark-1.3-mclaude-fable-5-1: 1,334 (Elo (human preference)) 1,334claude-fable-5-1
Chart data as a table
Top rows on Design Arena (All categories)
ModelLabElo (human preference)
claude-opus-5-5Anthropic1,389
gpt-6-astraOpenAI1,383
muse-wonder-1.0Meta1,377
gpt-6-astra-maxOpenAI1,371
kimi-k3Chinese labs1,359
gpt-6.1-solOpenAI1,358
claude-sonnet-5-5Anthropic1,354
muse-spark-1.3Meta1,350
muse-spark-1.3-maxMeta1,349
claude-fable-5-1Anthropic1,334

Source: designarena.ai

Share of top-10 rowsShare of top-10 rows on Design Arena (All categories), as of 2026-10-24. Counts rows, not a score.
OpenAI: 3 of 10Anthropic: 3 of 10Meta: 3 of 10Chinese labs: 1 of 103/10OpenAI
Share of top-10 rows on Design Arena (All categories), as of 2026-10-24. Counts rows, not a score.
LabRows
OpenAI3
Anthropic3
Meta3
Chinese labs1

Source: designarena.ai

  • New row on the board since the Oct 19 refresh
  • Design Arena per-category boards (Website, UI Component, Mobile Apps, Native Apps) require a Bearer API key per Design Arena API docs (https://docs.designarena.ai/api-reference/leaderboard); only the server-rendered All-categories and Fullstack boards were captured.
  • Design Arena per-category boards (android, mobile, ui-components) still need an API key
Design Arena (All categories), Elo (human preference), as of 2026-10-24. Values copied as published.
Board #Model (as listed)EloDetailLicenseAs ofSource
1claude-opus-5-5Anthropic · latest gen 1389 6,496 battlesNot listed ↗
2gpt-6-astraOpenAI · latest gen 1383 10,786 battlesNot listed ↗
3muse-wonder-1.0Meta · latest gen 1377 7,604 battlesNot listed ↗
4gpt-6-astra-maxOpenAI · latest gen 1371 3,831 battlesNot listed ↗
5kimi-k3Moonshot · latest gen 1359 9,056 battlesOpen weightskimi-k3 (custom) ↗
6gpt-6.1-solOpenAI · latest gen 1358 5,276 battlesNot listed ↗
7claude-sonnet-5-5Anthropic · latest gen 1354 4,735 battlesNot listed ↗
8muse-spark-1.3Meta · latest gen 1350 46,492 battlesNot listed ↗
9muse-spark-1.3-maxMeta · latest gen 1349 39,572 battlesNot listed ↗
10gpt-5.6-sol-xhighOpenAI · older version 1340 13,755 battlesNot listed ↗
11claude-fable-5-1Anthropic · latest gen 1334 9,349 battlesNot listed ↗
12gemini-4-argonGoogle · latest gen 1329 12,965 battlesNot listed ↗
13claude-opus-5Anthropic · older version 1322 19,481 battlesNot listed ↗
14deepseek-v4-1-flashDeepSeek · latest gen 1318 42,235 battlesOpen weightsmit ↗
15hampshire 1315 35,308 battlesNot listed ↗
16gpt-5.6-solOpenAI · older version 1315 28,417 battlesNot listed ↗
17muse-spark-1.2Meta · older version 1312 50,568 battlesNot listed ↗
18claude-fable-5Anthropic · older version 1309 20,217 battlesNot listed ↗
19gemini-3.7-flashGoogle · older version 1308 77,618 battlesNot listed ↗
20colorado 1304 15,128 battlesNot listed ↗
21gemini-3.8-flashGoogle · latest gen 1302 46,277 battlesNot listed ↗
22cortado 1301 21,626 battlesNot listed ↗
24gpt-6-solOpenAI · older version 1300 5,663 battlesNot listed ↗
24deepseek-v4-1-flash-1017DeepSeek · latest gen 1296 4,402 battlesNot listed ↗
26qwen3.8-max-previewAlibaba · latest gen 1292 11,396 battlesNot listed ↗
27grogu 1290 36,822 battlesNot listed ↗
28gemini-3.6-flashGoogle · older version 1289 32,832 battlesNot listed ↗
29claude-opus-4-6Anthropic · older version 1288 27,406 battlesNot listed ↗
30glm-5.2Z.ai 1286 57,124 battlesOpen weightsmit ↗
31claude-opus-4-6-thinkingAnthropic · older version 1284 23,814 battlesNot listed ↗
32notebook 1280 28,728 battlesNot listed ↗
33claude-haiku-5-5Anthropic · latest gen 1277 9,553 battlesNot listed ↗
34deca-2.6-ultra 1276 2,450 battlesNot listed ↗
35claude-sonnet-5Anthropic · older version 1275 24,519 battlesNot listed ↗
36evergreen 1275 23,956 battlesNot listed ↗
36claude-sonnet-4-6Anthropic · older version 1275 26,317 battlesNot listed ↗
38qwen3.7-maxAlibaba 1274 36,725 battlesNot listed ↗
39yoda 1274 37,544 battlesNot listed ↗
40grok-4.7SpaceXAI · latest gen 1271 3,025 battlesNot listed ↗
41kimi-k2.6Moonshot · older version 1271 34,048 battlesOpen weightsmodified-mit ↗
42deca-2.5-mini 1270 4,212 battlesNot listed ↗
43gpt-6-lunaOpenAI · latest gen 1269 7,483 battlesNot listed ↗
47mimo-v2.5-proXiaomi 1265 28,909 battlesOpen weightsmit ↗
51mimo-v2.5Xiaomi 1259 44,973 battlesOpen weightsmit ↗
58minimax-m3MiniMax · latest gen 1245 27,874 battlesOpen weightsminimax-community ↗
62deepseek-v4-proDeepSeek 1239 27,277 battlesOpen weightsmit ↗
70deepseek-v4-flash-0731DeepSeek 1228 10,439 battlesOpen weightsmit ↗
73grok-4-20-beta-reasoningSpaceXAI · older version 1220 37,055 battlesNot listed ↗
79grok-4-20-beta-non-reasoningSpaceXAI · older version 1205 40,222 battlesNot listed ↗
79deepseek-v4-flashDeepSeek 1204 28,104 battlesOpen weightsmit ↗
83grok-4.3SpaceXAI · older version 1186 41,413 battlesNot listed ↗
118nemotron-3-ultra-550b-a55b 1145 26,124 battlesOpen weightsopenmdw-1.1 ↗
140grok-4-fast-non-reasoningSpaceXAI · older version 1100 36,650 battlesNot listed ↗
143grok-4-1-fast-non-reasoningSpaceXAI · older version 1091 33,603 battlesNot listed ↗
144grok-4-1-fast-reasoningSpaceXAI · older version 1086 31,135 battlesNot listed ↗
153grok-3SpaceXAI · older version 1052 26,869 battlesNot listed ↗
155grok-4-fast-reasoningSpaceXAI · older version 1050 37,376 battlesNot listed ↗
163grok-4SpaceXAI · older version 1015 24,016 battlesNot listed ↗
168grok-code-fast-1SpaceXAI · older version 998 4,299 battlesNot listed ↗
173gpt-oss-120bOpenAI 961 5,272 battlesOpen weightsapache-2.0 ↗
176grok-3-miniSpaceXAI · older version 930 7,621 battlesNot listed ↗
177llama-3.1-nemotron-ultra-253b 928 3,173 battlesNot listed ↗
179llama-4-maverick 878 1,678 battlesOpen weightsllama4 (custom) ↗
183devstral-small-2507Mistral 806 1,250 battlesOpen weightsapache-2.0 ↗
184llama-4-scout 788 1,275 battlesOpen weightsllama4 (custom) ↗

Copied from designarena.ai/leaderboard · checked 2026-10-24 · Board # is the board's own rank

Browse models → · How scores are handled

The weekly diff

What changed in AI coding plans: prices, limits, resets. Email signups open soon. Until then, updates go out on X.

Follow @QuietlyByG on X RSS feed

No email form yet, so we don't collect any email addresses.