Skip to main content

Human ratings

English hard prompts (human ratings)

A category computed only from the English questions within the hard questions category.

How to read this table

  • Higher rows are rated higher. A range like “1–2” means the gap is within the margin of error.
  • The rating comes from people comparing two answers and picking the better one.
  • At a gap of 50, the higher model is picked about 57% of the time; 100 means 64%, 200 means 76%.
  • Prices are a guide per million tokens, reading then writing.
  • Length is how much text the model can read at once (tokens).

± range of the top model

RankModelDeveloperratingrating scale 1450 to 1600Week
change
OverallInput $
/1M
Output $
/1M
ContextSupportsWeightsReleasedPrice and specs
1–9Gemini 4 ArgonGoogleRating 1538 ±13−141517rank 1–3$2$101.05MGoogle Released 2026-⁠09 Week change −14 Price $2 / $10 Length 1.05M
1–10Claude Opus 5.5AnthropicRating 1537 ±13+81509rank 1–7$4$201MAnthropic Released 2026-⁠09 Week change +8 Price $4 / $20 Length 1M
1–8Claude Opus 4.6AnthropicRating 1534 ±5−21503rank 2–9$5$251MAnthropic Released 2026-⁠02 Week change −2 Price $5 / $25 Length 1M
1–10Claude Fable 5AnthropicRating 1531 ±6−11503rank 2–9$10$501MAnthropic Released 2026-⁠06 Week change −1 Price $10 / $50 Length 1M
1–12Claude Opus 4.7AnthropicRating 1528 ±601501rank 2–10$5$251MAnthropic Released 2026-⁠04 Price $5 / $25 Length 1M
1–26Muse WonderMetaRating 1523 ±16−11509rank 1–8$4$201.05MMeta Released 2026-⁠10 Week change −1 Price $4 / $20 Length 1.05M
1–25Claude Fable 5.1AnthropicRating 1521 ±10+51503rank 2–10$10$501MAnthropic Released 2026-⁠09 Week change +5 Price $10 / $50 Length 1M
5–26Kimi K3⁠MoonshotRating 1517 ±7+11487rank 8–27$3$151.05MMoonshot Released 2026-⁠07 Week change +1 Price $3 / $15 Length 1.05M
6–24Claude Opus 4.8AnthropicRating 1517 ±601481rank 12–36$5$251MAnthropic Released 2026-⁠05 Price $5 / $25 Length 1M
3–26Muse Spark 1.3MetaRating 1516 ±9+41496rank 3–15$1.25$4.251.05MMeta Released 2026-⁠09 Week change +4 Price $1.25 / $4.25 Length 1.05M
6–26Gemini 3.8 FlashGoogleRating 1514 ±7+11496rank 4–15$0.75$3.751.05MGoogle Released 2026-⁠09 Week change +1 Price $0.75 / $3.75 Length 1.05M
6–26Claude Opus 5AnthropicRating 1514 ±601491rank 8–22$5$251MAnthropic Released 2026-⁠07 Price $5 / $25 Length 1M
1–56MiMo-V2.6 Pro⁠XiaomiRating 1513 ±1901480rank 10–44$0.435$0.871.05MXiaomi Released 2026-⁠09 Price $0.435 / $0.87 Length 1.05M
2–40Claude Sonnet 5.5AnthropicRating 1513 ±1301478rank 14–44$2$101MAnthropic Released 2026-⁠09 Price $2 / $10 Length 1M
6–37GLM-5.3⁠Z.aiRating 1511 ±8+11481rank 11–38$1.4$4.41MZ.ai Released 2026-⁠08 Week change +1 Price $1.4 / $4.4 Length 1M
6–37Qwen3.8 MaxAlibabaRating 1510 ±801483rank 10–36$2$61MAlibaba Released 2026-⁠08 Price $2 / $6 Length 1M
5–53GPT-6.1 SolOpenAIRating 1510 ±14−31480rank 10–41$2$101.05MOpenAI Released 2026-⁠09 Week change −3 Price $2 / $10 Length 1.05M
6–49Muse Spark 1.2MetaRating 1509 ±12−11491rank 6–25$1.25$4.251.05MMeta Released 2026-⁠08 Week change −1 Price $1.25 / $4.25 Length 1.05M
6–37GPT-5.6 SolOpenAIRating 1509 ±701485rank 10–30$4$201.05MOpenAI Released 2026-⁠07 Price $4 / $20 Length 1.05M
6–44Muse SparkMetaRating 1508 ±901489rank 8–27———Meta Released 2026-⁠04 Price — Length —
6–48GPT-6 AstraOpenAIRating 1508 ±11+31476rank 17–44$10$501.05MOpenAI Released 2026-⁠09 Week change +3 Price $10 / $50 Length 1.05M
6–37Gemini 3.1 Pro (preview)GoogleRating 1507 ±401487rank 10–26$2$121.05MGoogle Released 2026-⁠02 Price $2 / $12 Length 1.05M
6–40Claude Sonnet 4.6AnthropicRating 1506 ±501472rank 24–49$3$151MAnthropic Released 2026-⁠02 Price $3 / $15 Length 1M
7–46Muse Spark 1.1MetaRating 1505 ±601490rank 8–23$1.25$4.251MMeta Released 2026-⁠07 Price $1.25 / $4.25 Length 1M
6–58DeepSeek-V4.1 Flash⁠DeepSeekRating 1504 ±1101476rank 17–46——1MDeepSeek Released 2026-⁠09 Price — Length 1M
8–51Claude Opus 4.5AnthropicRating 1503 ±701474rank 23–46$5$25200KAnthropic Released 2025-⁠11 Price $5 / $25 Length 200K
13–53MiMo-V2.5 Pro⁠XiaomiRating 1500 ±501466rank 34–53$0.435$0.871.05MXiaomi Released 2026-⁠04 Price $0.435 / $0.87 Length 1.05M
13–54GPT-5.5OpenAIRating 1500 ±501482rank 12–36$5$301.05MOpenAI Released 2026-⁠04 Price $5 / $30 Length 1.05M
13–59Gemini 3.6 FlashGoogleRating 1500 ±701483rank 10–35$0.75$3.751.05MGoogle Released 2026-⁠07 Price $0.75 / $3.75 Length 1.05M
13–59GPT-5.2 ChatOpenAIRating 1500 ±701476rank 19–42——128KOpenAI Released 2026-⁠02 Price — Length 128K
13–59Gemini 3 ProGoogleRating 1500 ±701486rank 10–29——1.05MGoogle Released 2025-⁠11 Price — Length 1.05M
13–59Gemini 3.7 FlashGoogleRating 1499 ±8+21490rank 8–25$0.75$3.751.05MGoogle Released 2026-⁠08 Week change +2 Price $0.75 / $3.75 Length 1.05M
13–57GLM-5.1⁠Z.aiRating 1499 ±501465rank 36–53$1.4$4.4200KZ.ai Released 2026-⁠04 Price $1.4 / $4.4 Length 200K
13–59GLM-5.3 Flash⁠Z.aiRating 1499 ±7−11476rank 19–43$0.15$0.51MZ.ai Released 2026-⁠08 Week change −1 Price $0.15 / $0.5 Length 1M
13–59GLM-5.2⁠Z.aiRating 1498 ±6+11475rank 19–43$1.4$4.41MZ.ai Released 2026-⁠06 Week change +1 Price $1.4 / $4.4 Length 1M
19–61Claude Sonnet 4.5AnthropicRating 1494 ±501456rank 49–69$3$15200KAnthropic Released 2025-⁠09 Price $3 / $15 Length 200K
17–68Qwen3.5 Max (preview)AlibabaRating 1494 ±801465rank 36–60————Alibaba Released — Price — Length —
19–65Claude Sonnet 5AnthropicRating 1494 ±601461rank 41–64$2$101MAnthropic Released 2026-⁠06 Price $2 / $10 Length 1M
20–66GPT-5.4OpenAIRating 1493 ±601475rank 19–42$2.5$151.05MOpenAI Released 2026-⁠03 Price $2.5 / $15 Length 1.05M
19–67Grok 4.5xAIRating 1493 ±701466rank 33–55$2$6500KxAI Released 2026-⁠07 Price $2 / $6 Length 500K
How to read this page
  • These ratings exclude the effect of response length and formatting. Differences where the ± ranges overlap are within the margin of error, so about the same.
  • In the Supports column the icons are, from left: can use tools / can read images / can answer in a fixed format (JSON); faded means not supported or unknown.
  • Open-weights models have their contents (the weights) published, so you can also run them on your own computer or server. Models without the open-weights mark are normally used through the developer's own services (apps or API); for some models the weights status is unknown.
  • Names such as ChatGPT next to a model are the consumer services its developer offers. Check on the service itself whether that model is available there.

Data: Arena, Epoch AI, Swallow LLM team (CC BY 4.0), Models.dev (MIT), updated 2026-⁠10-⁠24, data sources