Skip to main content
Bench

Language

English日本語中文Español

Site info

Task results

AIME-style math competition problems (OTIS Mock AIME 2024-2025)

Problems from unofficial mock exams modeled on the AIME, a US math exam, written by students of the OTIS math program.

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 score is the share of a shared set of problems the model solved (%).
  • ± is how far the score moves when re-measured, so small gaps do not mean much.
  • Prices are a guide per million tokens, reading then writing.
  • Length is how much text the model can read at once (tokens).
RankModelDeveloperAccuracy
(%)Accuracy scale 85 to 100		EvaluatedInput $
/1M		Output $
/1M		ContextSupportsWeightsReleasedPrice and specs
1–42Claude Fable 5.1AnthropicRating 100.0 ±0.0$10$501MAnthropic Released 2026-⁠09 Price $10 / $50 Length 1M
1–42Claude Fable 5AnthropicRating 100.0 ±0.0$10$501MAnthropic Released 2026-⁠06 Price $10 / $50 Length 1M
1–42Claude Opus 5.5AnthropicRating 100.0 ±0.0$4$201MAnthropic Released 2026-⁠09 Price $4 / $20 Length 1M
1–42Claude Sonnet 5.5AnthropicRating 100.0 ±0.0$2$101MAnthropic Released 2026-⁠09 Price $2 / $10 Length 1M
1–42Gemini 4 ArgonGoogleRating 100.0 ±0.0$2$101.05MGoogle Released 2026-⁠09 Price $2 / $10 Length 1.05M
1–42GPT-5.5OpenAIRating 100.0 ±0.0$5$301.05MOpenAI Released 2026-⁠04 Price $5 / $30 Length 1.05M
1–42GPT-5.5 ProOpenAIRating 100.0 ±0.0$30$1801.05MOpenAI Released 2026-⁠04 Price $30 / $180 Length 1.05M
1–42GPT-5.6 SolOpenAIRating 100.0 ±0.0$4$201.05MOpenAI Released 2026-⁠07 Price $4 / $20 Length 1.05M
1–42GPT-6.1 SolOpenAIRating 100.0 ±0.0$2$101.05MOpenAI Released 2026-⁠09 Price $2 / $10 Length 1.05M
1–42GPT-6 AstraOpenAIRating 100.0 ±0.0$10$501.05MOpenAI Released 2026-⁠09 Price $10 / $50 Length 1.05M
1–42GPT-6 SolOpenAIRating 100.0 ±0.0$2$101.05MOpenAI Released 2026-⁠09 Price $2 / $10 Length 1.05M
1–42Qwen3.8 Max (2026-⁠09)AlibabaRating 100.0 ±0.0——1MAlibaba Released 2026-⁠09 Price — Length 1M
1–45GPT-5.6 TerraOpenAIRating 99.7 ±0.5$2$121.05MOpenAI Released 2026-⁠07 Price $2 / $12 Length 1.05M
1–50DeepSeek-V4.1 Flash (2026-⁠10)⁠DeepSeekRating 99.4 ±0.8$0.15$0.61MDeepSeek Released 2026-⁠10 Price $0.15 / $0.6 Length 1M
1–50Muse WonderMetaRating 99.4 ±0.8$4$201.05MMeta Released 2026-⁠10 Price $4 / $20 Length 1.05M
1–50Qwen3.8 MaxAlibabaRating 99.4 ±0.8$2$61MAlibaba Released 2026-⁠08 Price $2 / $6 Length 1M
1–52Grok 4.6xAIRating 99.2 ±0.9$2$6500KxAI Released 2026-⁠08 Price $2 / $6 Length 500K
1–52Muse Spark 1.3MetaRating 99.2 ±1.2$1.25$4.251.05MMeta Released 2026-⁠09 Price $1.25 / $4.25 Length 1.05M
13–52Claude Haiku 5.5AnthropicRating 98.9 ±1.1$0.1$0.51MAnthropic Released 2026-⁠10 Price $0.1 / $0.5 Length 1M
1–54Claude Opus 5AnthropicRating 98.9 ±2.2$5$251MAnthropic Released 2026-⁠07 Price $5 / $25 Length 1M
1–52Gemini 3.8 FlashGoogleRating 98.9 ±1.7$0.75$3.751.05MGoogle Released 2026-⁠09 Price $0.75 / $3.75 Length 1.05M
1–52GPT-6 LunaOpenAIRating 98.9 ±1.3$0.1$0.51.05MOpenAI Released 2026-⁠09 Price $0.1 / $0.5 Length 1.05M
1–56DeepSeek-V4 Pro (2026-⁠08)⁠DeepSeekRating 98.6 ±1.9$0.66$1.981MDeepSeek Released 2026-⁠08 Price $0.66 / $1.98 Length 1M
1–61Claude Opus 4.8AnthropicRating 98.3 ±2.8$5$251MAnthropic Released 2026-⁠05 Price $5 / $25 Length 1M
14–56DeepSeek-V4.1 Flash⁠DeepSeekRating 98.3 ±1.3——1MDeepSeek Released 2026-⁠09 Price — Length 1M
1–60GPT-5.6 LunaOpenAIRating 98.3 ±2.3$0.2$1.21.05MOpenAI Released 2026-⁠07 Price $0.2 / $1.2 Length 1.05M
14–56Mistral Large 4MistralRating 98.3 ±1.3$1.36$4.18524.3KMistral Released 2026-⁠10 Price $1.36 / $4.18 Length 524.3K
1–63Grok 4.7xAIRating 98.1 ±2.8$2$6500KxAI Released 2026-⁠09 Price $2 / $6 Length 500K
1–62Muse Spark Open⁠MetaRating 98.1 ±2.5——1.04MMeta Released 2026-⁠10 Price — Length 1.04M
1–63Claude Opus 4.7AnthropicRating 97.8 ±4.3$5$251MAnthropic Released 2026-⁠04 Price $5 / $25 Length 1M
1–63GPT-5.4OpenAIRating 97.8 ±4.4$2.5$151.05MOpenAI Released 2026-⁠03 Price $2.5 / $15 Length 1.05M
1–63Grok 4.5xAIRating 97.8 ±2.5$2$6500KxAI Released 2026-⁠07 Price $2 / $6 Length 500K
1–70Gemini 3.7 FlashGoogleRating 97.2 ±3.6$0.75$3.751.05MGoogle Released 2026-⁠08 Price $0.75 / $3.75 Length 1.05M
14–63Kimi K3⁠MoonshotRating 97.2 ±2.2$3$151.05MMoonshot Released 2026-⁠07 Price $3 / $15 Length 1.05M
1–71DeepSeek-V4 Pro⁠DeepSeekRating 96.7 ±3.9$0.66$1.981MDeepSeek Released 2026-⁠04 Price $0.66 / $1.98 Length 1M
1–76GPT-5.2OpenAIRating 96.1 ±5.1$1.75$14400KOpenAI Released 2025-⁠12 Price $1.75 / $14 Length 400K
1–76Kimi K2.6⁠MoonshotRating 96.1 ±4.7$0.95$4262.1KMoonshot Released 2026-⁠04 Price $0.95 / $4 Length 262.1K
1–79Gemini 3.1 Pro (preview)GoogleRating 95.6 ±6.1$2$121.05MGoogle Released 2026-⁠02 Price $2 / $12 Length 1.05M
1–79Gemini 3.5 FlashGoogleRating 95.6 ±5.2$1.5$91.05MGoogle Released 2026-⁠05 Price $1.5 / $9 Length 1.05M
1–79Gemini 3 Flash (preview)GoogleRating 95.6 ±6.1$0.5$31MGoogle Released 2025-⁠12 Price $0.5 / $3 Length 1M

Best accuracy over time

2023-03-14Release date2026-10-17
How to read this page
  • Answers are integers, and the value is accuracy. They are placed as harder than MATH level 5 and easier than FrontierMath.
  • The accuracy is the share of problems solved, with the part you could get by guessing removed.
  • 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