SVG · Models
Design Arena's Elo for svg · models, from pairwise human preference votes, as OpenRouter carries it. 83 models ranked, as carried in OpenRouter's catalog.
retrieved sha256 · 0bb53a77713fwww.designarena.ai ↗
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
- 01DeepSeek V4 Flash 07311169 · $0.09 blended
- 02DeepSeek V4.1 Flash 10171241 · $0.152 blended · +72 at 1.7x
- 03GLM 5.3 Flash1269 · $0.238 blended · +28 at 1.6x
- 04MiMo-V2.6-Pro1290 · $0.544 blended · +21 at 2.3x
- 05Muse Spark 1.31314 · $2.00 blended · +24 at 3.7x
- 06Kimi K3.1 Pro1324 · $2.281 blended · +10 at 1.1x
- 07Gemini 4 Carbon1400 · $4.00 blended · +76 at 1.8x
83 of 83 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 83 priced models sit in the attractive quadrant: at or above the median Elo of 1170 and at or below the median blended price of $1.55. The tinted corner is that quadrant. 83 of the 83 crosses match the selection below and are drawn white.
83 of 83.
Design Arena · Elo3 more authors are on this board, the largest with 1 model. Search a name to reach one.
| Position | Model | Author | Elo | Win rate | Source rank | In / out per M | Context | Listed |
|---|---|---|---|---|---|---|---|---|
| 01 | Gemini 4 Carbon | 1400 | 69.3% | 1 | $2.00 / $10.00 | 1.05M | ||
| 02 | Gemini 4 Argon | 1339 | 62.2% | 2 | $2.00 / $10.00 | 1.05M | ||
| 03 | GPT-6.1 Sol | OpenAI | 1336 | 59.7% | 3 | $2.00 / $10.00 | 1.05M | |
| 04 | Kimi K3.1 Pro | MoonshotAI | 1324 | 61.9% | 4 | $0.24 / $8.40 | 1.05M | |
| 05 | Claude Opus 5.5 | Anthropic | 1321 | 54.0% | 5 | $4.00 / $20.00 | 1M | |
| 06 | Muse Spark 1.3 | Meta | 1314 | 55.2% | 6 | $1.25 / $4.25 | 1.05M | |
| 07 | Claude Opus 5 | Anthropic | 1305 | 58.5% | 7 | $5.00 / $25.00 | 1M | |
| 08 | Claude Sonnet 5.5 | Anthropic | 1300 | 56.0% | 8 | $2.00 / $10.00 | 1M | |
| 09 | Kimi K3 | MoonshotAI | 1296 | 59.5% | 9 | $0.192 / $16.00 | 1.05M | |
| 10 | Claude Fable 5 | Anthropic | 1292 | 62.8% | 10 | $10.00 / $50.00 | 1M | |
| 11 | Claude Fable 5.1 | Anthropic | 1292 | 56.6% | 11 | $10.00 / $50.00 | 1M | |
| 12 | MiMo-V2.6-Pro | Xiaomi | 1290 | 54.2% | 12 | $0.435 / $0.87 | 1.05M | |
| 13 | Gemini 3.8 Flash | 1289 | 54.8% | 13 | $0.825 / $4.13 | 1.05M | ||
| 14 | Gemini 3.7 Flash | 1276 | 53.8% | 14 | $0.75 / $3.75 | 1.05M | ||
| 15 | Gemini 3.1 Pro Preview | 1274 | 67.9% | 15 | $2.00 / $12.00 | 1.05M | ||
| 16 | Grok 4.7 | SpaceXAI | 1270 | 50.4% | 16 | $2.00 / $6.00 | 500k | |
| 17 | Muse Spark 1.2 | Meta | 1270 | 54.2% | 17 | $1.25 / $4.25 | 1.05M | |
| 18 | GLM 5.3 Flash | Z.ai | 1269 | 54.6% | 18 | $0.15 / $0.50 | 1.05M | |
| 19 | Gemini 3.5 Flash | 1245 | 58.0% | 19 | $1.50 / $9.00 | 1.05M | ||
| 20 | GLM 5.3 | Z.ai | 1244 | 51.1% | 20 | $0.0699 / $1.40 | 1.05M | |
| 21 | DeepSeek V4.1 Flash 1017 | DeepSeek | 1241 | 50.7% | 21 | $0.036 / $0.50 | 1.05M | |
| 22 | GPT-5.5 | OpenAI | 1225 | 54.7% | 22 | $5.00 / $30.00 | 1.05M | |
| 23 | Muse Spark 1.1 | Meta | 1223 | 49.1% | 23 | $1.25 / $4.25 | 1.05M | |
| 24 | Mistral Large 4 | Mistral | 1222 | 44.9% | 24 | $0.70 / $2.10 | 1.05M | |
| 25 | Grok 4.6 | SpaceXAI | 1221 | 47.8% | 25 | $2.00 / $6.00 | 500k | |
| 26 | Claude Opus 4.6 | Anthropic | 1217 | 60.9% | 26 | $5.00 / $25.00 | 1M | |
| 27 | GLM 5.1 | Z.ai | 1207 | 54.9% | 27 | $1.40 / $4.40 | 205k | |
| 28 | Qwen3.7 Max | Qwen | 1205 | 54.7% | 28 | $1.48 / $4.42 | 1M | |
| 29 | Grok 4.5 | SpaceXAI | 1200 | 47.5% | 29 | $2.00 / $6.00 | 500k | |
| 30 | Qwen3.7 Plus | Qwen | 1199 | 46.8% | 30 | $0.32 / $1.28 | 1M | |
| 31 | Claude Haiku 5.5 | Anthropic | 1196 | 43.9% | 31 | $0.10 / $0.50 | 1M | |
| 32 | GLM 5.2 | Z.ai | 1195 | 47.3% | 32 | $0.007 / $6.80 | 1.05M | |
| 33 | GLM 5 Turbo | Z.ai | 1194 | 53.8% | 33 | $1.20 / $4.00 | 203k | |
| 34 | Claude Sonnet 4.6 | Anthropic | 1186 | 58.7% | 34 | $3.00 / $15.00 | 1M | |
| 35 | GPT-5.4 | OpenAI | 1184 | 57.8% | 36 | $2.50 / $15.00 | 1.05M | |
| 36 | MiMo-V2.5-Pro | Xiaomi | 1184 | 48.8% | 37 | $0.435 / $0.87 | 1.02M | |
| 37 | Claude Sonnet 5 | Anthropic | 1183 | 48.8% | 38 | $2.00 / $10.00 | 1M | |
| 38 | Claude Opus 4.8 | Anthropic | 1181 | 50.4% | 39 | $5.00 / $25.00 | 1M | |
| 39 | GPT-5 | OpenAI | 1178 | 64.1% | 40 | $1.25 / $10.00 | 400k | |
| 40 | MiMo-V2.5 | Xiaomi | 1177 | 50.5% | 41 | $0.40 / $2.00 | 1M | |
| 41 | Claude Opus 4.5 | Anthropic | 1170 | 58.7% | 42 | $5.00 / $25.00 | 200k | |
| 42 | Kimi K2.6 | MoonshotAI | 1170 | 51.2% | 43 | $0.65 / $3.41 | 262k | |
| 43 | DeepSeek V4 Flash 0731 | DeepSeek | 1169 | 44.7% | 44 | $0.06 / $0.18 | 1.05M | |
| 44 | Kimi K2.7 Code | MoonshotAI | 1164 | 43.6% | 46 | $0.70 / $3.75 | 262k | |
| 45 | GLM 5 | Z.ai | 1157 | 54.4% | 47 | $1.00 / $3.20 | 205k | |
| 46 | Grok 4.20 | SpaceXAI | 1153 | 52.4% | 48 | $1.25 / $2.50 | 2M | |
| 47 | Qwen3.6 Plus | Qwen | 1152 | 50.0% | 49 | $0.325 / $1.95 | 1M | |
| 48 | MiniMax M3 | MiniMax | 1146 | 43.8% | 50 | $0.30 / $1.20 | 1.05M | |
| 49 | DeepSeek V4 Flash 0423 | DeepSeek | 1144 | 48.4% | 52 | $0.09 / $0.18 | 1.05M | |
| 50 | Claude Opus 4.1 | Anthropic | 1141 | 60.8% | 53 | $15.00 / $75.00 | 200k | |
| 51 | MiniMax M2.5 | MiniMax | 1138 | 54.5% | 54 | $0.295 / $1.20 | 198k | |
| 52 | GLM 5V Turbo | Z.ai | 1138 | 48.0% | 55 | $1.20 / $4.00 | 203k | |
| 53 | GPT-5.1 | OpenAI | 1138 | 57.4% | 56 | $1.25 / $10.00 | 400k | |
| 54 | Kimi K2.5 | MoonshotAI | 1138 | 48.4% | 57 | $0.50 / $2.50 | 262k | |
| 55 | GLM 4.7 | Z.ai | 1132 | 54.3% | 58 | $0.40 / $1.75 | 205k | |
| 56 | DeepSeek V4 Pro 0423 | DeepSeek | 1128 | 45.5% | 59 | $0.90 / $1.50 | 1.05M | |
| 57 | GPT-5.2 | OpenAI | 1128 | 53.7% | 60 | $1.75 / $14.00 | 400k | |
| 58 | Qwen3.5 397B A17B | Qwen | 1128 | 56.1% | 61 | $0.45 / $3.00 | 262k | |
| 59 | MiniMax M2.7 | MiniMax | 1123 | 46.5% | 62 | $0.21 / $0.84 | 197k | |
| 60 | GPT-5.3-Codex | OpenAI | 1118 | 54.0% | 64 | $1.75 / $14.00 | 400k | |
| 61 | Claude Sonnet 4.5 | Anthropic | 1104 | 52.2% | 66 | $3.00 / $15.00 | 1M | |
| 62 | Inkling | Thinking Machines | 1101 | 34.4% | 67 | $1.00 / $4.05 | 524k | |
| 63 | GLM 4.6 | Z.ai | 1098 | 52.1% | 68 | $0.50 / $2.00 | 205k | |
| 64 | Qwen3.5 Plus 2026-02-15 | Qwen | 1097 | 48.9% | 69 | $0.26 / $1.56 | 1M | |
| 65 | GLM 4.5 | Z.ai | 1085 | 50.8% | 71 | $0.60 / $2.20 | 131k | |
| 66 | GPT-5 Mini | OpenAI | 1080 | 45.8% | 73 | $0.25 / $2.00 | 400k | |
| 67 | Step 3.7 Flash | StepFun | 1072 | 36.2% | 75 | $0.20 / $1.15 | 262k | |
| 68 | Grok 4.3 | SpaceXAI | 1070 | 36.5% | 76 | $1.25 / $2.50 | 1M | |
| 69 | GLM 4.5 Air | Z.ai | 1060 | 50.8% | 79 | $0.13 / $0.85 | 131k | |
| 70 | Nemotron 3 Ultra | NVIDIA | 1060 | 32.0% | 80 | $0.50 / $2.20 | 262k | |
| 71 | DeepSeek V3.1 Terminus | DeepSeek | 1054 | 50.1% | 81 | $0.27 / $1.00 | 164k | |
| 72 | Gemini 3.1 Flash Lite Preview | 1041 | 42.5% | 83 | $0.25 / $1.50 | 1.05M | ||
| 73 | R1 0528 | DeepSeek | 1028 | 48.7% | 86 | $0.50 / $2.15 | 164k | |
| 74 | GLM 4.7 Flash | Z.ai | 1027 | 44.2% | 87 | $0.0605 / $0.40 | 200k | |
| 75 | DeepSeek V3.2 | DeepSeek | 1021 | 40.8% | 88 | $0.26 / $0.38 | 164k | |
| 76 | DeepSeek V3.2 Exp | DeepSeek | 1021 | 42.0% | 89 | $0.27 / $0.41 | 164k | |
| 77 | Claude Haiku 4.5 | Anthropic | 1014 | 39.1% | 90 | $1.00 / $5.00 | 200k | |
| 78 | Gemini 2.5 Flash | 1009 | 43.1% | 91 | $0.30 / $2.50 | 1.05M | ||
| 79 | Mistral Large 3 2512 | Mistral | 982 | 38.0% | 95 | $0.50 / $1.50 | 262k | |
| 80 | Mistral Medium 3.1 | Mistral | 982 | 38.2% | 96 | $0.44 / $2.20 | 131k | |
| 81 | GPT-5.1-Codex-Mini | OpenAI | 967 | 35.3% | 97 | $0.25 / $2.00 | 400k | |
| 82 | DeepSeek V3 | DeepSeek | 965 | 38.8% | 98 | $0.257 / $1.03 | 164k | |
| 83 | DeepSeek V3.1 | DeepSeek | 956 | 38.2% | 99 | $0.25 / $0.95 | 164k |
The field and the frontier above are the whole board and do not narrow with this table: a frontier drawn through 83 of 83 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.
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.
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.
