Get started

pdf · Agents

Design Arena's Elo for pdf · agents, from pairwise human preference votes, as OpenRouter carries it. 36 models ranked, as carried in OpenRouter's catalog.

retrieved sha256 · 0bb53a77713fwww.designarena.ai ↗

Most capable1351Claude Fable 5.2Anthropic
Best value6300Claude Haiku 5.51260 at $0.20 blended per M
Newest ranked1220Gemini 4 Hydrogenlisted 12 Nov 2026

5 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$1.00$10.0094311471351Blended list price per M tokens, log scaleEloClaude Fable 5.2: 1351 · $16.00 blended · Anthropic$16.00135105 Claude Fable 5.21351 · $16.00 blended · AnthropicClaude Opus 5.5: 1316 · $8.00 blended · Anthropic$8.00131604 Claude Opus 5.51316 · $8.00 blended · AnthropicGemini 4 Carbon: 1305 · $4.00 blended · Google$4.00130503 Gemini 4 Carbon1305 · $4.00 blended · GoogleGemini 4 Argon: 1282 · $4.00 blended · Google$4.001282Gemini 4 Argon1282 · $4.00 blended · GoogleKimi K3: 1279 · $4.144 blended · MoonshotAI$4.1441279Kimi K31279 · $4.144 blended · MoonshotAIMiniMax M3: 1267 · $0.525 blended · MiniMax$0.525126702 MiniMax M31267 · $0.525 blended · MiniMaxClaude Haiku 5.5: 1260 · $0.20 blended · Anthropic$0.20126001 Claude Haiku 5.51260 · $0.20 blended · AnthropicGLM 5.2: 1252 · $1.705 blended · Z.ai$1.7051252GLM 5.21252 · $1.705 blended · Z.aiClaude Opus 5: 1249 · $10.00 blended · Anthropic$10.001249Claude Opus 51249 · $10.00 blended · AnthropicClaude Sonnet 5.5: 1247 · $4.00 blended · Anthropic$4.001247Claude Sonnet 5.51247 · $4.00 blended · AnthropicMuse Wonder: 1234 · $8.00 blended · Meta$8.001234Muse Wonder1234 · $8.00 blended · MetaGrok 4.6: 1231 · $3.00 blended · SpaceXAI$3.001231Grok 4.61231 · $3.00 blended · SpaceXAIKimi K3.1 Pro: 1222 · $2.281 blended · MoonshotAI$2.2811222Kimi K3.1 Pro1222 · $2.281 blended · MoonshotAIClaude Sonnet 5: 1220 · $4.00 blended · Anthropic$4.001220Claude Sonnet 51220 · $4.00 blended · AnthropicGemini 4 Hydrogen: 1220 · $3.00 blended · Google$3.001220Gemini 4 Hydrogen1220 · $3.00 blended · GoogleGemini 3.8 Flash: 1210 · $1.65 blended · Google$1.651210Gemini 3.8 Flash1210 · $1.65 blended · GoogleMuse Spark 1.3: 1203 · $2.00 blended · Meta$2.001203Muse Spark 1.31203 · $2.00 blended · MetaGLM 5.1: 1200 · $2.15 blended · Z.ai$2.151200GLM 5.11200 · $2.15 blended · Z.aiGrok 4.7: 1197 · $3.00 blended · SpaceXAI$3.001197Grok 4.71197 · $3.00 blended · SpaceXAIClaude Opus 4.6: 1191 · $10.00 blended · Anthropic$10.001191Claude Opus 4.61191 · $10.00 blended · AnthropicClaude Opus 4.8: 1175 · $10.00 blended · Anthropic$10.001175Claude Opus 4.81175 · $10.00 blended · AnthropicGemini 3.6 Flash: 1174 · $1.50 blended · Google$1.501174Gemini 3.6 Flash1174 · $1.50 blended · GoogleQwen3.7 Max: 1165 · $2.213 blended · Qwen$2.2131165Qwen3.7 Max1165 · $2.213 blended · QwenMuse Spark 1.1: 1159 · $2.00 blended · Meta$2.001159Muse Spark 1.11159 · $2.00 blended · MetaGemini 3.7 Flash: 1156 · $1.50 blended · Google$1.501156Gemini 3.7 Flash1156 · $1.50 blended · GoogleStep 3.7 Flash: 1152 · $0.438 blended · StepFun$0.4381152Step 3.7 Flash1152 · $0.438 blended · StepFunGemini 3.5 Flash: 1148 · $3.375 blended · Google$3.3751148Gemini 3.5 Flash1148 · $3.375 blended · GoogleGPT-6.1 Sol: 1131 · $4.00 blended · OpenAI$4.001131GPT-6.1 Sol1131 · $4.00 blended · OpenAIGrok 4.5: 1128 · $3.00 blended · SpaceXAI$3.001128Grok 4.51128 · $3.00 blended · SpaceXAIMuse Spark 1.2: 1124 · $2.00 blended · Meta$2.001124Muse Spark 1.21124 · $2.00 blended · MetaGemini 3.1 Pro Preview: 1107 · $4.50 blended · Google$4.501107Gemini 3.1 Pro Preview1107 · $4.50 blended · GoogleMuse Glimmer 30B: 1107 · $0.638 blended · Meta$0.6381107Muse Glimmer 30B1107 · $0.638 blended · MetaQwen3.7 Plus: 1101 · $0.56 blended · Qwen$0.561101Qwen3.7 Plus1101 · $0.56 blended · QwenMuse Spark Open: 1039 · $2.00 blended · Meta$2.001039Muse Spark Open1039 · $2.00 blended · MetaNemotron 3.5 Super VL: 984 · $0.238 blended · NVIDIA$0.238984Nemotron 3.5 Super VL984 · $0.238 blended · NVIDIAGrok 4.3: 943 · $1.563 blended · SpaceXAI$1.563943Grok 4.3943 · $1.563 blended · SpaceXAI
  1. 01Claude Haiku 5.51260 · $0.20 blended
  2. 02MiniMax M31267 · $0.525 blended · +7 at 2.6x
  3. 03Gemini 4 Carbon1305 · $4.00 blended · +38 at 7.6x
  4. 04Claude Opus 5.51316 · $8.00 blended · +11 at 2.0x
  5. 05Claude Fable 5.21351 · $16.00 blended · +35 at 2.0x

36 of 36 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. 7 of 36 priced models sit in the attractive quadrant: at or above the median Elo of 1198.5 and at or below the median blended price of $2.64. The tinted corner is that quadrant. 8 of the 36 crosses match the selection below and are drawn white.

8 of 36.

Design Arena · Elo
PositionModelAuthorEloWin rateSource rankIn / out per MContextListed
01Claude Fable 5.2Anthropic135172.5%1$8.00 / $40.001M
02Claude Opus 5.5Anthropic131668.0%2$4.00 / $20.001M
07Claude Haiku 5.5Anthropic126063.0%8$0.10 / $0.501M
09Claude Opus 5Anthropic124958.9%10$5.00 / $25.001M
10Claude Sonnet 5.5Anthropic124758.6%11$2.00 / $10.001M
14Claude Sonnet 5Anthropic122054.8%15$2.00 / $10.001M
20Claude Opus 4.6Anthropic119151.4%21$5.00 / $25.001M
21Claude Opus 4.8Anthropic117550.3%22$5.00 / $25.001M

The field and the frontier above are the whole board and do not narrow with this table: a frontier drawn through 8 of 36 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.