Get started

ASCII art · Models

Design Arena's Elo for ascii art · models, from pairwise human preference votes, as OpenRouter carries it. 67 models ranked, as carried in OpenRouter's catalog.

retrieved sha256 · 0bb53a77713fwww.designarena.ai ↗

Most capable1458Gemini 4 HydrogenGoogle
Best value12022DeepSeek V4 Flash 07311082 at $0.09 blended per M
Newest ranked1458Gemini 4 Hydrogenlisted 12 Nov 2026

7 models sit on the frontier: nothing scores higher for less.

  • the attractive quadrant
  • the frontier
  • a priced model, in its maker's tone
  • on the frontier, named
The attractive quadrant$0.10$1.00$10.0098112201458Blended list price per M tokens, log scaleEloGemini 4 Hydrogen: 1458 · $3.00 blended · Google$3.00145807 Gemini 4 Hydrogen1458 · $3.00 blended · GoogleKimi K3: 1351 · $4.144 blended · MoonshotAI$4.1441351Kimi K31351 · $4.144 blended · MoonshotAIClaude Opus 5.5: 1349 · $8.00 blended · Anthropic$8.001349Claude Opus 5.51349 · $8.00 blended · AnthropicClaude Opus 5: 1338 · $10.00 blended · Anthropic$10.001338Claude Opus 51338 · $10.00 blended · AnthropicMuse Spark 1.3: 1320 · $2.00 blended · Meta$2.00132006 Muse Spark 1.31320 · $2.00 blended · MetaClaude Fable 5: 1310 · $20.00 blended · Anthropic$20.001310Claude Fable 51310 · $20.00 blended · AnthropicClaude Fable 5.1: 1307 · $20.00 blended · Anthropic$20.001307Claude Fable 5.11307 · $20.00 blended · AnthropicKimi K3.1 Pro: 1296 · $2.281 blended · MoonshotAI$2.2811296Kimi K3.1 Pro1296 · $2.281 blended · MoonshotAIGemini 3.6 Flash: 1269 · $1.50 blended · Google$1.50126905 Gemini 3.6 Flash1269 · $1.50 blended · GoogleMuse Spark 1.2: 1266 · $2.00 blended · Meta$2.001266Muse Spark 1.21266 · $2.00 blended · MetaClaude Opus 4.8: 1260 · $10.00 blended · Anthropic$10.001260Claude Opus 4.81260 · $10.00 blended · AnthropicGemini 3.8 Flash: 1259 · $1.65 blended · Google$1.651259Gemini 3.8 Flash1259 · $1.65 blended · GoogleGemini 3.7 Flash: 1258 · $1.50 blended · Google$1.501258Gemini 3.7 Flash1258 · $1.50 blended · GoogleGLM 5.3 Flash: 1257 · $0.238 blended · Z.ai$0.238125704 GLM 5.3 Flash1257 · $0.238 blended · Z.aiGrok 4.6: 1256 · $3.00 blended · SpaceXAI$3.001256Grok 4.61256 · $3.00 blended · SpaceXAIGemini 3.1 Pro Preview: 1253 · $4.50 blended · Google$4.501253Gemini 3.1 Pro Preview1253 · $4.50 blended · GoogleMuse Spark 1.1: 1244 · $2.00 blended · Meta$2.001244Muse Spark 1.11244 · $2.00 blended · MetaClaude Opus 4.6: 1238 · $10.00 blended · Anthropic$10.001238Claude Opus 4.61238 · $10.00 blended · AnthropicGemini 3.5 Flash: 1237 · $3.375 blended · Google$3.3751237Gemini 3.5 Flash1237 · $3.375 blended · GoogleGrok 4.5: 1237 · $3.00 blended · SpaceXAI$3.001237Grok 4.51237 · $3.00 blended · SpaceXAIGPT-5.5: 1235 · $11.25 blended · OpenAI$11.251235GPT-5.51235 · $11.25 blended · OpenAIClaude Sonnet 4.6: 1219 · $6.00 blended · Anthropic$6.001219Claude Sonnet 4.61219 · $6.00 blended · AnthropicDeepSeek V4.1 Flash 1017: 1219 · $0.152 blended · DeepSeek$0.152121903 DeepSeek V4.1 Flash 10171219 · $0.152 blended · DeepSeekClaude Haiku 5.5: 1204 · $0.20 blended · Anthropic$0.201204Claude Haiku 5.51204 · $0.20 blended · AnthropicMistral Large 4: 1193 · $1.05 blended · Mistral$1.051193Mistral Large 41193 · $1.05 blended · MistralQwen3.7 Max: 1193 · $2.213 blended · Qwen$2.2131193Qwen3.7 Max1193 · $2.213 blended · QwenClaude Sonnet 5: 1189 · $4.00 blended · Anthropic$4.001189Claude Sonnet 51189 · $4.00 blended · AnthropicClaude Sonnet 4.5: 1188 · $6.00 blended · Anthropic$6.001188Claude Sonnet 4.51188 · $6.00 blended · AnthropicGLM 5.3: 1184 · $0.402 blended · Z.ai$0.4021184GLM 5.31184 · $0.402 blended · Z.aiGPT-5.4: 1182 · $5.625 blended · OpenAI$5.6251182GPT-5.41182 · $5.625 blended · OpenAIClaude Opus 4.5: 1174 · $10.00 blended · Anthropic$10.001174Claude Opus 4.51174 · $10.00 blended · AnthropicGLM 5.2: 1174 · $1.705 blended · Z.ai$1.7051174GLM 5.21174 · $1.705 blended · Z.aiKimi K2.7 Code: 1174 · $1.463 blended · MoonshotAI$1.4631174Kimi K2.7 Code1174 · $1.463 blended · MoonshotAIGrok 4.20: 1158 · $1.563 blended · SpaceXAI$1.5631158Grok 4.201158 · $1.563 blended · SpaceXAIClaude Opus 4.1: 1150 · $30.00 blended · Anthropic$30.001150Claude Opus 4.11150 · $30.00 blended · AnthropicGemini 3.1 Flash Lite Preview: 1150 · $0.563 blended · Google$0.5631150Gemini 3.1 Flash Lite Preview1150 · $0.563 blended · GoogleKimi K2.5: 1149 · $1.00 blended · MoonshotAI$1.001149Kimi K2.51149 · $1.00 blended · MoonshotAIGLM 4.7: 1148 · $0.738 blended · Z.ai$0.7381148GLM 4.71148 · $0.738 blended · Z.aiGLM 5 Turbo: 1146 · $1.90 blended · Z.ai$1.901146GLM 5 Turbo1146 · $1.90 blended · Z.aiMiMo-V2.5-Pro: 1142 · $0.544 blended · Xiaomi$0.5441142MiMo-V2.5-Pro1142 · $0.544 blended · XiaomiKimi K2.6: 1141 · $1.34 blended · MoonshotAI$1.341141Kimi K2.61141 · $1.34 blended · MoonshotAIMiniMax M3: 1139 · $0.525 blended · MiniMax$0.5251139MiniMax M31139 · $0.525 blended · MiniMaxGPT-5.2: 1136 · $4.813 blended · OpenAI$4.8131136GPT-5.21136 · $4.813 blended · OpenAIGPT-5.3-Codex: 1135 · $4.813 blended · OpenAI$4.8131135GPT-5.3-Codex1135 · $4.813 blended · OpenAIDeepSeek V4 Pro 0423: 1132 · $1.05 blended · DeepSeek$1.051132DeepSeek V4 Pro 04231132 · $1.05 blended · DeepSeekStep 3.7 Flash: 1132 · $0.438 blended · StepFun$0.4381132Step 3.7 Flash1132 · $0.438 blended · StepFunGLM 5: 1130 · $1.55 blended · Z.ai$1.551130GLM 51130 · $1.55 blended · Z.aiGLM 5.1: 1129 · $2.15 blended · Z.ai$2.151129GLM 5.11129 · $2.15 blended · Z.aiMiMo-V2.5: 1129 · $0.80 blended · Xiaomi$0.801129MiMo-V2.51129 · $0.80 blended · XiaomiQwen3.7 Plus: 1125 · $0.56 blended · Qwen$0.561125Qwen3.7 Plus1125 · $0.56 blended · QwenGPT-5: 1124 · $3.438 blended · OpenAI$3.4381124GPT-51124 · $3.438 blended · OpenAIClaude Haiku 4.5: 1123 · $2.00 blended · Anthropic$2.001123Claude Haiku 4.51123 · $2.00 blended · AnthropicMiniMax M2.7: 1118 · $0.368 blended · MiniMax$0.3681118MiniMax M2.71118 · $0.368 blended · MiniMaxGrok 4.3: 1109 · $1.563 blended · SpaceXAI$1.5631109Grok 4.31109 · $1.563 blended · SpaceXAIGPT-5 Mini: 1108 · $0.688 blended · OpenAI$0.6881108GPT-5 Mini1108 · $0.688 blended · OpenAIGLM 5V Turbo: 1103 · $1.90 blended · Z.ai$1.901103GLM 5V Turbo1103 · $1.90 blended · Z.aiGPT-5.1: 1102 · $3.438 blended · OpenAI$3.4381102GPT-5.11102 · $3.438 blended · OpenAIQwen3.6 Plus: 1096 · $0.731 blended · Qwen$0.7311096Qwen3.6 Plus1096 · $0.731 blended · QwenDeepSeek V4 Flash 0423: 1093 · $0.113 blended · DeepSeek$0.113109302 DeepSeek V4 Flash 04231093 · $0.113 blended · DeepSeekGPT-5.1-Codex-Mini: 1088 · $0.688 blended · OpenAI$0.6881088GPT-5.1-Codex-Mini1088 · $0.688 blended · OpenAIDeepSeek V4 Flash 0731: 1082 · $0.09 blended · DeepSeek$0.09108201 DeepSeek V4 Flash 07311082 · $0.09 blended · DeepSeekInkling: 1082 · $1.763 blended · Thinking Machines$1.7631082Inkling1082 · $1.763 blended · Thinking MachinesNemotron 3 Ultra: 1079 · $0.925 blended · NVIDIA$0.9251079Nemotron 3 Ultra1079 · $0.925 blended · NVIDIAQwen3.5 Plus 2026-02-15: 1077 · $0.585 blended · Qwen$0.5851077Qwen3.5 Plus 2026-02-151077 · $0.585 blended · QwenDeepSeek V3.2: 1066 · $0.29 blended · DeepSeek$0.291066DeepSeek V3.21066 · $0.29 blended · DeepSeekMistral Large 3 2512: 1053 · $0.75 blended · Mistral$0.751053Mistral Large 3 25121053 · $0.75 blended · MistralMistral Medium 3.1: 981 · $0.88 blended · Mistral$0.88981Mistral Medium 3.1981 · $0.88 blended · Mistral
  1. 01DeepSeek V4 Flash 07311082 · $0.09 blended
  2. 02DeepSeek V4 Flash 04231093 · $0.113 blended · +11 at 1.3x
  3. 03DeepSeek V4.1 Flash 10171219 · $0.152 blended · +126 at 1.4x
  4. 04GLM 5.3 Flash1257 · $0.238 blended · +38 at 1.6x
  5. 05Gemini 3.6 Flash1269 · $1.50 blended · +12 at 6.3x
  6. 06Muse Spark 1.31320 · $2.00 blended · +51 at 1.3x
  7. 07Gemini 4 Hydrogen1458 · $3.00 blended · +138 at 1.5x

67 of 67 ranked models carry a list price; 0 are free or unlisted and not in the field. Price is the blended OpenRouter rate, three input tokens to one output, on a logarithmic axis. Each dot is filled in its maker's tone, as the columns above are. The numbered and named dots are the frontier, cheapest first; each step after the first names what it adds over the step before, and at what multiple of the blended price. A hover over any dot names it and its two figures, a click opens the model, and a model named in the picker above keeps its label on. 11 of 67 priced models sit in the attractive quadrant: at or above the median Elo of 1158 and at or below the median blended price of $1.705. The tinted corner is that quadrant. 67 of the 67 crosses match the selection below and are drawn white.

67 of 67.

Design Arena · Elo
PositionModelAuthorEloWin rateSource rankIn / out per MContextListed
01Gemini 4 HydrogenGoogle145874.1%1$1.50 / $7.501.05M
02Kimi K3MoonshotAI135168.3%2$0.192 / $16.001.05M
03Claude Opus 5.5Anthropic134964.8%3$4.00 / $20.001M
04Claude Opus 5Anthropic133867.6%4$5.00 / $25.001M
05Muse Spark 1.3Meta132062.5%5$1.25 / $4.251.05M
06Claude Fable 5Anthropic131067.9%6$10.00 / $50.001M
07Claude Fable 5.1Anthropic130765.0%7$10.00 / $50.001M
08Kimi K3.1 ProMoonshotAI129658.5%8$0.24 / $8.401.05M
09Gemini 3.6 FlashGoogle126957.6%9$0.75 / $3.751.05M
10Muse Spark 1.2Meta126654.6%10$1.25 / $4.251.05M
11Claude Opus 4.8Anthropic126061.3%11$5.00 / $25.001M
12Gemini 3.8 FlashGoogle125951.9%12$0.825 / $4.131.05M
13Gemini 3.7 FlashGoogle125853.3%13$0.75 / $3.751.05M
14GLM 5.3 FlashZ.ai125753.9%14$0.15 / $0.501.05M
15Grok 4.6SpaceXAI125656.6%15$2.00 / $6.00500k
16Gemini 3.1 Pro PreviewGoogle125363.5%16$2.00 / $12.001.05M
17Muse Spark 1.1Meta124452.9%17$1.25 / $4.251.05M
18Claude Opus 4.6Anthropic123861.9%18$5.00 / $25.001M
19Gemini 3.5 FlashGoogle123758.8%19$1.50 / $9.001.05M
20Grok 4.5SpaceXAI123752.6%20$2.00 / $6.00500k
21GPT-5.5OpenAI123557.2%21$5.00 / $30.001.05M
22Claude Sonnet 4.6Anthropic121960.1%22$3.00 / $15.001M
23DeepSeek V4.1 Flash 1017DeepSeek121947.8%23$0.036 / $0.501.05M
24Claude Haiku 5.5Anthropic120449.6%24$0.10 / $0.501M
25Mistral Large 4Mistral119343.5%25$0.70 / $2.101.05M
26Qwen3.7 MaxQwen119353.0%26$1.48 / $4.421M
27Claude Sonnet 5Anthropic118950.8%27$2.00 / $10.001M
28Claude Sonnet 4.5Anthropic118856.1%28$3.00 / $15.001M
29GLM 5.3Z.ai118445.1%29$0.0699 / $1.401.05M
30GPT-5.4OpenAI118255.4%30$2.50 / $15.001.05M
31Claude Opus 4.5Anthropic117454.8%32$5.00 / $25.00200k
32GLM 5.2Z.ai117446.3%33$0.007 / $6.801.05M
33Kimi K2.7 CodeMoonshotAI117447.7%34$0.70 / $3.75262k
34Grok 4.20SpaceXAI115848.3%35$1.25 / $2.502M
35Claude Opus 4.1Anthropic115051.5%38$15.00 / $75.00200k
36Gemini 3.1 Flash Lite PreviewGoogle115050.8%39$0.25 / $1.501.05M
37Kimi K2.5MoonshotAI114946.5%40$0.50 / $2.50262k
38GLM 4.7Z.ai114848.2%41$0.40 / $1.75205k
39GLM 5 TurboZ.ai114648.5%42$1.20 / $4.00203k
40MiMo-V2.5-ProXiaomi114246.8%43$0.435 / $0.871.02M
41Kimi K2.6MoonshotAI114146.1%44$0.65 / $3.41262k
42MiniMax M3MiniMax113945.4%45$0.30 / $1.201.05M
43GPT-5.2OpenAI113650.7%46$1.75 / $14.00400k
44GPT-5.3-CodexOpenAI113551.2%47$1.75 / $14.00400k
45DeepSeek V4 Pro 0423DeepSeek113246.4%48$0.90 / $1.501.05M
46Step 3.7 FlashStepFun113244.3%49$0.20 / $1.15262k
47GLM 5Z.ai113047.9%50$1.00 / $3.20205k
48GLM 5.1Z.ai112944.2%51$1.40 / $4.40205k
49MiMo-V2.5Xiaomi112945.1%52$0.40 / $2.001M
50Qwen3.7 PlusQwen112540.7%53$0.32 / $1.281M
51GPT-5OpenAI112449.0%54$1.25 / $10.00400k
52Claude Haiku 4.5Anthropic112349.3%55$1.00 / $5.00200k
53MiniMax M2.7MiniMax111845.9%56$0.21 / $0.84197k
54Grok 4.3SpaceXAI110942.1%58$1.25 / $2.501M
55GPT-5 MiniOpenAI110844.8%59$0.25 / $2.00400k
56GLM 5V TurboZ.ai110342.4%60$1.20 / $4.00203k
57GPT-5.1OpenAI110248.6%61$1.25 / $10.00400k
58Qwen3.6 PlusQwen109640.6%62$0.325 / $1.951M
59DeepSeek V4 Flash 0423DeepSeek109342.7%63$0.09 / $0.181.05M
60GPT-5.1-Codex-MiniOpenAI108843.0%64$0.25 / $2.00400k
61DeepSeek V4 Flash 0731DeepSeek108234.0%66$0.06 / $0.181.05M
62InklingThinking Machines108234.3%67$1.00 / $4.05524k
63Nemotron 3 UltraNVIDIA107936.3%68$0.50 / $2.20262k
64Qwen3.5 Plus 2026-02-15Qwen107743.2%69$0.26 / $1.561M
65DeepSeek V3.2DeepSeek106640.5%70$0.26 / $0.38164k
66Mistral Large 3 2512Mistral105340.3%72$0.50 / $1.50262k
67Mistral Medium 3.1Mistral98130.8%75$0.44 / $2.20131k

The field and the frontier above are the whole board and do not narrow with this table: a frontier drawn through 67 of 67 models would say nothing beats them on both while something in the same board does. The models this table holds are drawn white in the figures instead.

What the score tells you

An Elo from pairwise human preference votes: a higher figure means the model's output was preferred more often against the field it was matched with. Win rate is the share of matchups won. Both shift as new models enter the arena, neither is a measure of correctness, and the source's own rank counts models this catalog does not carry.

Sources

Source: OpenRouter (openrouter.ai/models), as of 17 Nov 16:29 UTC. sha256 0bb53a77713f over 853,270 bytes.
Catalog: OpenRouter's public model list.
Intelligence, coding and agentic indexes: Artificial Analysis, as published by OpenRouter.
Elo, win rate and rank: Design Arena, as published by OpenRouter.
Vitna ran none of these evaluations and adjusted none of these figures.