Skip to main content

Human ratings

Math (human ratings)

A category computed only from well-defined math problems solved by calculation or logical deduction.

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 1400 to 1600Week
change
OverallInput $
/1M
Output $
/1M
ContextSupportsWeightsReleasedPrice and specs
1–27Gemini 4 ArgonGoogleRating 1532 ±24+31517rank 1–3$2$101.05MGoogle Released 2026-⁠09 Week change +3 Price $2 / $10 Length 1.05M
1–26Claude Fable 5.1AnthropicRating 1529 ±21+51503rank 2–10$10$501MAnthropic Released 2026-⁠09 Week change +5 Price $10 / $50 Length 1M
1–22Gemini 3.8 FlashGoogleRating 1526 ±13+31496rank 4–15$0.75$3.751.05MGoogle Released 2026-⁠09 Week change +3 Price $0.75 / $3.75 Length 1.05M
1–27Claude Fable 5AnthropicRating 1521 ±13−31503rank 2–9$10$501MAnthropic Released 2026-⁠06 Week change −3 Price $10 / $50 Length 1M
1–24Claude Opus 5AnthropicRating 1521 ±11−21491rank 8–22$5$251MAnthropic Released 2026-⁠07 Week change −2 Price $5 / $25 Length 1M
1–27Claude Opus 4.6AnthropicRating 1517 ±10−11503rank 2–9$5$251MAnthropic Released 2026-⁠02 Week change −1 Price $5 / $25 Length 1M
1–47GPT-6.1 SolOpenAIRating 1513 ±26+221480rank 10–41$2$101.05MOpenAI Released 2026-⁠09 Week change +22 Price $2 / $10 Length 1.05M
1–45Claude Opus 5.5AnthropicRating 1512 ±25+91509rank 1–7$4$201MAnthropic Released 2026-⁠09 Week change +9 Price $4 / $20 Length 1M
1–40Muse Spark 1.3MetaRating 1510 ±18+61496rank 3–15$1.25$4.251.05MMeta Released 2026-⁠09 Week change +6 Price $1.25 / $4.25 Length 1.05M
1–42Gemini 3.7 FlashGoogleRating 1505 ±15−11490rank 8–25$0.75$3.751.05MGoogle Released 2026-⁠08 Week change −1 Price $0.75 / $3.75 Length 1.05M
1–42Gemini 3.6 FlashGoogleRating 1504 ±14+11483rank 10–35$0.75$3.751.05MGoogle Released 2026-⁠07 Week change +1 Price $0.75 / $3.75 Length 1.05M
1–37Claude Opus 4.7AnthropicRating 1504 ±1101501rank 2–10$5$251MAnthropic Released 2026-⁠04 Price $5 / $25 Length 1M
1–42GLM-5.3 Flash⁠Z.aiRating 1504 ±14−21476rank 19–43$0.15$0.51MZ.ai Released 2026-⁠08 Week change −2 Price $0.15 / $0.5 Length 1M
1–60Claude Sonnet 5.5AnthropicRating 1501 ±27−91478rank 14–44$2$101MAnthropic Released 2026-⁠09 Week change −9 Price $2 / $10 Length 1M
2–42GPT-5.5OpenAIRating 1501 ±1001482rank 12–36$5$301.05MOpenAI Released 2026-⁠04 Price $5 / $30 Length 1.05M
1–55GLM-5.3⁠Z.aiRating 1500 ±17+41481rank 11–38$1.4$4.41MZ.ai Released 2026-⁠08 Week change +4 Price $1.4 / $4.4 Length 1M
1–54Kimi K3⁠MoonshotRating 1499 ±15−11487rank 8–27$3$151.05MMoonshot Released 2026-⁠07 Week change −1 Price $3 / $15 Length 1.05M
1–55Qwen3.8 MaxAlibabaRating 1499 ±15−11483rank 10–36$2$61MAlibaba Released 2026-⁠08 Week change −1 Price $2 / $6 Length 1M
2–51Gemini 3.5 FlashGoogleRating 1497 ±12+11477rank 18–41$1.5$91.05MGoogle Released 2026-⁠05 Week change +1 Price $1.5 / $9 Length 1.05M
3–55GPT-5.6 SolOpenAIRating 1496 ±1301485rank 10–30$4$201.05MOpenAI Released 2026-⁠07 Price $4 / $20 Length 1.05M
1–72Muse WonderMetaRating 1496 ±31New1509rank 1–8$4$201.05MMeta Released 2026-⁠10 Price $4 / $20 Length 1.05M
1–60DeepSeek-V4.1 Flash⁠DeepSeekRating 1495 ±2101476rank 17–46——1MDeepSeek Released 2026-⁠09 Price — Length 1M
7–55Claude Opus 4.8AnthropicRating 1494 ±1101481rank 12–36$5$251MAnthropic Released 2026-⁠05 Price $5 / $25 Length 1M
7–56Muse Spark 1.1MetaRating 1493 ±13−11490rank 8–23$1.25$4.251MMeta Released 2026-⁠07 Week change −1 Price $1.25 / $4.25 Length 1M
7–55GPT-5.4OpenAIRating 1491 ±1101475rank 19–42$2.5$151.05MOpenAI Released 2026-⁠03 Price $2.5 / $15 Length 1.05M
1–68Muse Spark 1.2MetaRating 1491 ±25+81491rank 6–25$1.25$4.251.05MMeta Released 2026-⁠08 Week change +8 Price $1.25 / $4.25 Length 1.05M
7–55Gemini 3.1 Pro (preview)GoogleRating 1489 ±801487rank 10–26$2$121.05MGoogle Released 2026-⁠02 Price $2 / $12 Length 1.05M
1–99MiMo-V2.6 Pro⁠XiaomiRating 1488 ±40+31480rank 10–44$0.435$0.871.05MXiaomi Released 2026-⁠09 Week change +3 Price $0.435 / $0.87 Length 1.05M
7–59GLM-5.2⁠Z.aiRating 1488 ±12+31475rank 19–43$1.4$4.41MZ.ai Released 2026-⁠06 Week change +3 Price $1.4 / $4.4 Length 1M
4–68GPT-6 AstraOpenAIRating 1487 ±21+51476rank 17–44$10$501.05MOpenAI Released 2026-⁠09 Week change +5 Price $10 / $50 Length 1.05M
1–106Qwen3.7 Max (preview)AlibabaRating 1484 ±38−21474rank 15–53————Alibaba Released — Week change −2 Price — Length —
7–76Tencent HY3⁠TencentRating 1484 ±22+61454rank 48–76——256KTencent Released 2026-⁠07 Week change +6 Price — Length 256K
7–62GPT-5.6 TerraOpenAIRating 1483 ±13+21467rank 28–53$2$121.05MOpenAI Released 2026-⁠07 Week change +2 Price $2 / $12 Length 1.05M
8–68Inkling⁠Thinking MachinesRating 1480 ±13+11442rank 68–97$1.87$4.68262.1KThinking Machines Released 2026-⁠07 Week change +1 Price $1.87 / $4.68 Length 262.1K
13–63MiMo-V2.5 Pro⁠XiaomiRating 1479 ±10−11466rank 34–53$0.435$0.871.05MXiaomi Released 2026-⁠04 Week change −1 Price $0.435 / $0.87 Length 1.05M
8–68Kimi K2.6⁠MoonshotRating 1479 ±13−11461rank 41–65$0.95$4262.1KMoonshot Released 2026-⁠04 Week change −1 Price $0.95 / $4 Length 262.1K
8–72GPT-5.6 LunaOpenAIRating 1478 ±1401453rank 51–75$0.2$1.21.05MOpenAI Released 2026-⁠07 Price $0.2 / $1.2 Length 1.05M
9–72ERNIE 5.1BaiduRating 1478 ±13−11467rank 26–53——119K—Baidu Released — Week change −1 Price — Length 119K
13–68Gemini 3 ProGoogleRating 1477 ±1101486rank 10–29——1.05MGoogle Released 2025-⁠11 Price — Length 1.05M
15–71GLM-5.1⁠Z.aiRating 1476 ±1101465rank 36–53$1.4$4.4200KZ.ai Released 2026-⁠04 Price $1.4 / $4.4 Length 200K
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