Best AI for design

Design here means producing an interface a person prefers, not writing about design.

Rumeqo runs these models inside your team rooms. See what each one costs.

available on rumeqo

115 of 115 ranked models

Ranked models
rank	model	vendor	composite	pricein / out	benchmarks	elo	elo	elo	elo
1	Claude Opus 5 (max)anthropic/claude-opus-5:max	Anthropic	82.4	$5.00 / $25.00	4 of 4 benchmarks	1691	1711	1679	1670
2	Qwen3.8 Maxqwen/qwen3.8-max	Qwen	80.6	$2.00 / $6.00	4 of 4 benchmarks	1669	1679	1628	1631
3	Claude Fable 5anthropic/claude-fable-5	Anthropic	80.0	$10.00 / $50.00	4 of 4 benchmarks	1627	1630	1671	1626
4	Kimi K3 (max)moonshotai/kimi-k3:max	MoonshotAI	79.2	$3.00 / $15.00	4 of 4 benchmarks	1674	1692	1650	1570
5	GPT-5.6 Sol (xhigh)openai/gpt-5.6-sol:xhigh	OpenAI	78.6	$5.00 / $30.00	4 of 4 benchmarks	1622	1629	1623	1581
6	Claude Opus 5 (high)anthropic/claude-opus-5:high	Anthropic	77.6	$5.00 / $25.00	3 of 4 benchmarks	1664	1681	1661
7	Grok 4.5x-ai/grok-4.5	SpaceXAI	75.9	$2.00 / $6.00	4 of 4 benchmarks	1553	1555	1578	1580
8	Grok 4.6 (high)x-ai/grok-4.6:high	SpaceXAI	75.8	$2.00 / $6.00	3 of 4 benchmarks	1618	1632	1619
9	Claude Opus 4.7anthropic/claude-opus-4.7	Anthropic	74.7	$5.00 / $25.00	4 of 4 benchmarks	1558	1554	1562	1565
10	Claude Opus 4.8 (high)anthropic/claude-opus-4.8:high	Anthropic	73.5	$5.00 / $25.00	3 of 4 benchmarks	1564	1562	1557
11	GLM 5.2 (max)z-ai/glm-5.2:max	Z.ai	73.2	$0.63 / $1.98	3 of 4 benchmarks	1587	1597	1540
12	Claude Opus 4.7 (high)anthropic/claude-opus-4.7:high	Anthropic	72.8	$5.00 / $25.00	3 of 4 benchmarks	1557	1558	1554
13	DeepSeek V4 Flash 0423 (high)deepseek/deepseek-v4-flash:high	DeepSeek	72.1	$0.14 / $0.28	3 of 4 benchmarks	1582	1592	1538
14	Claude Opus 4.6 (high)anthropic/claude-opus-4.6:high	Anthropic	71.7	$5.00 / $25.00	3 of 4 benchmarks	1545	1538	1559
15	Muse Spark 1.1meta/muse-spark-1.1	Meta	71.1	$1.25 / $4.25	4 of 4 benchmarks	1539	1530	1539	1544
16	Claude Sonnet 5 (high)anthropic/claude-sonnet-5:high	Anthropic	70.0	$2.00 / $10.00	4 of 4 benchmarks	1541	1546	1523	1540
17	Claude Opus 4.6anthropic/claude-opus-4.6	Anthropic	69.8	$5.00 / $25.00	4 of 4 benchmarks	1537	1529	1550	1538
18	Claude Opus 4.8anthropic/claude-opus-4.8	Anthropic	68.4	$5.00 / $25.00	3 of 4 benchmarks	1539	1542	1525
19	Muse Spark 1.2 (xhigh)meta/muse-spark-1.2:xhigh	Meta	68.1	$1.25 / $4.25	3 of 4 benchmarks	1535	1530	1539
20	Gemini 3.6 Flash (high)google/gemini-3.6-flash:high	Google	67.7	$1.50 / $7.50	3 of 4 benchmarks	1537	1530	1533
21	GPT-5.6 Terra (xhigh)openai/gpt-5.6-terra:xhigh	OpenAI	66.8	$1.00 / $6.00	4 of 4 benchmarks	1523	1514	1542	1525
22	Claude Sonnet 4.6anthropic/claude-sonnet-4.6	Anthropic	66.7	$3.00 / $15.00	4 of 4 benchmarks	1524	1513	1529	1530
23	GPT-5.5 (xhigh)openai/gpt-5.5:xhigh	OpenAI	65.5	$5.00 / $30.00	4 of 4 benchmarks	1509	1500	1544	1526
24	Qwen3.7 Maxqwen/qwen3.7-max	Qwen	64.5	$1.475 / $4.425	3 of 4 benchmarks	1517	1515	1521
25	Hy3tencent/hy3	Tencent	64.1	$0.132 / $0.528	3 of 4 benchmarks	1523	1536	1464
26	GLM 5.1z-ai/glm-5.1	Z.ai	63.7	$1.40 / $4.40	3 of 4 benchmarks	1511	1506	1521
27	Seed 2.1 Pro Previewbytedance/seed-2.1-pro-preview	ByteDance	62.7	—	4 of 4 benchmarks	1522	1524	1497	1506
28	GPT-5.6 Luna (xhigh)openai/gpt-5.6-luna:xhigh	OpenAI	62.6	$0.10 / $0.60	4 of 4 benchmarks	1518	1507	1537	1493
29	Kimi K2.6moonshotai/kimi-k2.6	MoonshotAI	62.0	$0.95 / $4.00	4 of 4 benchmarks	1509	1504	1495	1520
30	Gemini 3.5 Flash (high)google/gemini-3.5-flash:high	Google	61.6	$1.50 / $9.00	4 of 4 benchmarks	1506	1498	1541	1492
31	GPT-5.5 (high)openai/gpt-5.5:high	OpenAI	61.6	$5.00 / $30.00	4 of 4 benchmarks	1486	1477	1539	1511
32	Claude Opus 4.5 (high 32K)anthropic/claude-opus-4.5:high-32k	Anthropic	61.3	$5.00 / $25.00	3 of 4 benchmarks	1494	1480	1512
33	Claude Opus 4.7 (thinking)anthropic/claude-opus-4.7:thinking	Anthropic	60.8	$5.00 / $25.00	1 of 4 benchmarks				1579
34	Qwen3.6 Max Previewqwen/qwen3.6-max-preview	Qwen	59.4	$1.027 / $6.162	3 of 4 benchmarks	1479	1470	1495
35	Gemini 3.5 Flash (medium)google/gemini-3.5-flash:medium	Google	59.1	$1.50 / $9.00	4 of 4 benchmarks	1488	1482	1526	1490
36	MiMo-V2.5-Proxiaomi/mimo-v2.5-pro	Xiaomi	58.3	$0.435 / $0.87	3 of 4 benchmarks	1474	1462	1482
37	Claude Opus 4.5anthropic/claude-opus-4.5	Anthropic	57.9	$5.00 / $25.00	3 of 4 benchmarks	1468	1457	1482
38	Claude Opus 4.6 (thinking)anthropic/claude-opus-4.6:thinking	Anthropic	57.5	$5.00 / $25.00	1 of 4 benchmarks				1540
39	Kimi K2.7 Codemoonshotai/kimi-k2.7-code	MoonshotAI	57.4	$0.67 / $3.40	4 of 4 benchmarks	1473	1473	1462	1512
40	DeepSeek V4 Pro (high)deepseek/deepseek-v4-pro:high	DeepSeek	56.7	$1.168 / $2.336	3 of 4 benchmarks	1464	1455	1473
41	GPT-5.4 (high)openai/gpt-5.4:high	OpenAI	56.5	$2.50 / $15.00	3 of 4 benchmarks	1463	1453	1477
42	GPT-5.5openai/gpt-5.5	OpenAI	56.2	$5.00 / $30.00	4 of 4 benchmarks	1458	1446	1504	1499
43	MiniMax M3minimax/minimax-m3	MiniMax	55.8	$0.30 / $1.20	4 of 4 benchmarks	1491	1487	1464	1475
44	Gemini 3.6 Flashgoogle/gemini-3.6-flash	Google	54.3	$1.50 / $7.50	1 of 4 benchmarks				1529
45	GPT-5.4 (medium)openai/gpt-5.4:medium	OpenAI	54.2	$2.50 / $15.00	3 of 4 benchmarks	1442	1429	1474
46	Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview	Google	53.3	$2.00 / $12.00	4 of 4 benchmarks	1447	1437	1494	1481
47	DeepSeek V4 Prodeepseek/deepseek-v4-pro	DeepSeek	53.0	$1.168 / $2.336	3 of 4 benchmarks	1445	1439	1436
48	Qwen3.6 Plusqwen/qwen3.6-plus	Qwen	52.1	$0.325 / $1.95	4 of 4 benchmarks	1459	1454	1461	1470
49	GLM 4.7z-ai/glm-4.7	Z.ai	51.2	$0.40 / $1.75	3 of 4 benchmarks	1434	1429	1446
50	GLM 5z-ai/glm-5	Z.ai	50.7	$0.95 / $2.55	3 of 4 benchmarks	1436	1425	1454
51	MiMo-V2.5xiaomi/mimo-v2.5	Xiaomi	50.4	$0.14 / $0.28	3 of 4 benchmarks	1438	1427	1428
52	Mimo v2 Proxiaomi/mimo-v2-pro	Xiaomi	50.2	—	3 of 4 benchmarks	1434	1426	1439
53	GPT-5 (medium)openai/gpt-5:medium	OpenAI	49.9	$1.25 / $10.00	2 of 4 benchmarks	1419		1429
54	Gemini 3 Flash Previewgoogle/gemini-3-flash-preview	Google	49.5	$0.50 / $3.00	4 of 4 benchmarks	1438	1430	1444	1455
55	GPT-5.2openai/gpt-5.2	OpenAI	49.4	$1.75 / $14.00	2 of 4 benchmarks	1418		1428
56	Gemini 3 Progoogle/gemini-3-pro	Google	47.7	—	4 of 4 benchmarks	1438	1396	1469	1448
57	Kimi K2.5 (thinking)moonshotai/kimi-k2.5:thinking	MoonshotAI	46.6	$0.57 / $2.85	4 of 4 benchmarks	1436	1429	1429	1441
58	Qwen3.5 397B A17Bqwen/qwen3.5-397b-a17b	Qwen	45.6	$0.50 / $3.60	3 of 4 benchmarks	1400	1389	1413
59	GPT-5.1 (medium)openai/gpt-5.1:medium	OpenAI	45.3	$1.25 / $10.00	2 of 4 benchmarks	1391		1401
60	GPT-5.4 Mini (high)openai/gpt-5.4-mini:high	OpenAI	45.2	$0.75 / $4.50	3 of 4 benchmarks	1397	1387	1423
61	Inklingthinkingmachines/inkling	Thinking Machines	44.8	$0.95 / $4.05	3 of 4 benchmarks	1405	1399	1381
62	MiniMax M2.7minimax/minimax-m2.7	MiniMax	44.6	$0.30 / $1.20	3 of 4 benchmarks	1398	1388	1402
63	Gemini 3.5 Flash Litegoogle/gemini-3.5-flash-lite	Google	44.5	$0.30 / $2.50	3 of 4 benchmarks	1449	1426		1426
64	GPT-5.3-Codexopenai/gpt-5.3-codex	OpenAI	44.2	$1.75 / $14.00	4 of 4 benchmarks	1409	1398	1428	1446
65	Claude Opus 4.1anthropic/claude-opus-4.1	Anthropic	44.2	$15.00 / $75.00	2 of 4 benchmarks	1389		1398
66	Claude Sonnet 4.5 (high 32K)anthropic/claude-sonnet-4.5:high-32k	Anthropic	43.3	$3.00 / $15.00	3 of 4 benchmarks	1392	1386	1399
67	GPT-5.4openai/gpt-5.4	OpenAI	42.7	$2.50 / $15.00	4 of 4 benchmarks	1390	1388	1416	1449
68	MiniMax M2.5minimax/minimax-m2.5	MiniMax	42.5	$0.22 / $0.90	3 of 4 benchmarks	1384	1373	1415
69	MiniMax M2.1minimax/minimax-m2.1	MiniMax	41.7	$0.30 / $1.20	3 of 4 benchmarks	1387	1369	1401
70	Claude Sonnet 4.5anthropic/claude-sonnet-4.5	Anthropic	41.0	$3.00 / $15.00	3 of 4 benchmarks	1386	1384	1390
71	Kimi K2.5 Instantmoonshotai/kimi-k2.5-instant	Moonshot AI	40.5	—	4 of 4 benchmarks	1405	1392	1414	1418
72	Solar Pro 4upstage/solar-pro4	Upstage	39.6	$0.03 / $0.12	3 of 4 benchmarks	1372	1358	1394
73	Grok 4.20 Beta 0309 (reasoning)x-ai/grok-4.20-beta-0309:reasoning	xAI	38.6	—	3 of 4 benchmarks	1374	1367	1367
74	Gemma 4 31Bgoogle/gemma-4-31b-it	Google	38.3	$0.10 / $0.34	3 of 4 benchmarks	1365	1358	1378
75	Gemini 3 Flash Preview (thinking minimal)google/gemini-3-flash-preview:thinking-minimal	Google	37.8	$0.50 / $3.00	4 of 4 benchmarks	1383	1374	1398	1432
76	Gemma 4 26B A4B google/gemma-4-26b-a4b-it	Google	37.7	$0.12 / $0.40	3 of 4 benchmarks	1362	1354	1372
77	Muse Glimmer 30Bmeta/muse-glimmer-30b	Meta	37.5	$0.35 / $1.50	3 of 4 benchmarks	1359	1352	1385
78	GLM 5V Turboz-ai/glm-5v-turbo	Z.ai	37.4	$1.20 / $4.00	4 of 4 benchmarks	1400	1385	1365	1423
79	DeepSeek V3.2 (thinking)deepseek/deepseek-v3.2:thinking	DeepSeek	36.7	$0.269 / $0.40	3 of 4 benchmarks	1361	1350	1369
80	Qwen3.5-27Bqwen/qwen3.5-27b	Qwen	36.6	$0.195 / $1.56	3 of 4 benchmarks	1357	1345	1393
81	GLM 4.6z-ai/glm-4.6	Z.ai	36.4	$0.50 / $2.00	2 of 4 benchmarks	1340		1350
82	GPT-5.1 (high)openai/gpt-5.1:high	OpenAI	36.4	$1.25 / $10.00	1 of 4 benchmarks				1420
83	GPT-5.1-Codexopenai/gpt-5.1-codex	OpenAI	35.5	$1.25 / $10.00	2 of 4 benchmarks	1336		1346
84	Qwen3.5-122B-A10Bqwen/qwen3.5-122b-a10b	Qwen	35.0	$0.29 / $2.40	3 of 4 benchmarks	1358	1350	1357
85	Hunyuan Hy3 Previewtencent/hunyuan-hy3-preview	Tencent	34.7	—	3 of 4 benchmarks	1356	1348	1360
86	MiniMax M2minimax/minimax-m2	MiniMax	32.8	$0.255 / $1.02	2 of 4 benchmarks	1297		1307
87	Laguna M.1poolside/laguna-m.1	Poolside	32.4	—	3 of 4 benchmarks	1348	1339	1324
88	GPT-5.2-Codexopenai/gpt-5.2-codex	OpenAI	32.2	$1.75 / $14.00	3 of 4 benchmarks	1338	1330	1348
89	Mimo v2 Flashxiaomi/mimo-v2-flash	Xiaomi	31.5	—	3 of 4 benchmarks	1330	1310	1353
90	Grok 4.3x-ai/grok-4.3	SpaceXAI	31.2	$1.25 / $2.50	4 of 4 benchmarks	1355	1352	1360	1375
91	DeepSeek V3.2 Expdeepseek/deepseek-v3.2-exp	DeepSeek	30.9	$0.27 / $0.41	2 of 4 benchmarks	1272		1282
92	Claude Haiku 4.5anthropic/claude-haiku-4.5	Anthropic	30.3	$1.00 / $5.00	3 of 4 benchmarks	1326	1323	1316
93	Kimi K2 Thinking Turbomoonshotai/kimi-k2-thinking-turbo	Moonshot AI	30.3	—	3 of 4 benchmarks	1323	1317	1327
94	DeepSeek V3.2deepseek/deepseek-v3.2	DeepSeek	30.2	$0.269 / $0.40	3 of 4 benchmarks	1324	1335	1304
95	KAT-Coder-Pro V1kwaipilot/kat-coder-pro-v1	Kwaipilot	29.8	—	2 of 4 benchmarks	1255		1265
96	GPT-5.1-Codex-Miniopenai/gpt-5.1-codex-mini	OpenAI	28.9	$0.25 / $2.00	2 of 4 benchmarks	1244		1254
97	GPT-5.1openai/gpt-5.1	OpenAI	28.1	$1.25 / $10.00	4 of 4 benchmarks	1341	1305	1364	1345
98	Mimo v2 Flash (thinking)xiaomi/mimo-v2-flash:thinking	Xiaomi	28.0	—	3 of 4 benchmarks	1293	1231	1346
99	Mistral Medium 3.5mistralai/mistral-medium-3-5	Mistral	27.8	$1.50 / $7.50	3 of 4 benchmarks	1266	1255	1317
100	Laguna XS.2poolside/laguna-xs.2	Poolside	27.6	—	3 of 4 benchmarks	1303	1296	1281
101	Qwen3 Coder 480B A35b Instructqwen/qwen3-coder-480b-a35b-instruct	Qwen	27.2	—	3 of 4 benchmarks	1273	1257	1286
102	Grok 4.1 (thinking)x-ai/grok-4.1:thinking	xAI	26.4	—	2 of 4 benchmarks	1210		1220
103	Qwen3.5-35B-A3Bqwen/qwen3.5-35b-a3b	Qwen	26.1	$0.25 / $1.25	3 of 4 benchmarks	1250	1234	1298
104	Grok Code Fast 1x-ai/grok-code-fast-1	xAI	24.5	—	2 of 4 benchmarks	1164		1172
105	Qwen3.5-Flashqwen/qwen3.5-flash-02-23	Qwen	24.5	$0.065 / $0.26	3 of 4 benchmarks	1238	1224	1293
106	Grok 4 Fast (reasoning)x-ai/grok-4-fast:reasoning	xAI	24.1	—	2 of 4 benchmarks	1161		1170
107	Trinity Large Thinkingarcee-ai/trinity-large-thinking	Arcee AI	23.6	$0.22 / $0.85	3 of 4 benchmarks	1239	1233	1241
108	Devstral Medium 2507mistralai/devstral-medium-2507	Mistral AI	23.6	—	2 of 4 benchmarks	1080		1087
109	Grok 4.1 Fast (reasoning)x-ai/grok-4-1-fast:reasoning	xAI	23.4	—	3 of 4 benchmarks	1240	1222	1252
110	Mistral Large 3mistralai/mistral-large-3	Mistral AI	23.3	—	3 of 4 benchmarks	1230		1239	1361
111	Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview	Google	21.7	$0.25 / $1.50	4 of 4 benchmarks	1254	1242	1280	1336
112	Gemini 2.5 Progoogle/gemini-2.5-pro	Google	21.4	$1.25 / $10.00	3 of 4 benchmarks	1226		1235	1283
113	Granite 4.1 8Bibm-granite/granite-4.1-8b	IBM	21.0	$0.05 / $0.10	3 of 4 benchmarks	1192	1179	1219
114	Devstral 2mistralai/devstral-2	Mistral AI	20.5	—	3 of 4 benchmarks	1194	1139	1216
115	Mercury 2inception/mercury-2	Inception	20.2	$0.25 / $0.75	3 of 4 benchmarks	1166	1156	1199

How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 2 of the 4 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

Data sources

Turn a source off to drop every benchmark it feeds and rank the board again from what is left, in your browser. Turn them all off and the table has nothing to rank. Your choice follows you across the leaderboard pages.

LMArenaCC BY 4.0
Arena ratings by LMArena, from the public leaderboard dataset.