Skip to content
PAY.NEWS LEADERBOARD

AI model leaderboard

Evaluation sources How ranks are computed

Maths, logic and unfamiliar rules — can the model work out a problem it has not seen?

5 evaluations·2 operators·10/17 20:05 updated

Reasoning boardTOP 30

Ranked by the shared evidence of multiple public evaluations

Benchmark aggregation, not a payment-provider rating. Sponsorship, payment and affiliations must not affect results. Editorial standards

The top 30 models, sortable by column
Model
01gpt-6.1-sol—Evidence-sensitive · 5 evaluationsOfficial —99.0
02GPT-6 AstraOpenAIEvidence-sensitive · 5 evaluationsOfficial $10 / $5098.7
03Claude Opus 5.5AnthropicEvidence-sensitive · 5 evaluationsOfficial $4 / $2097.3
04Claude Fable 5.1AnthropicEvidence-sensitive · 5 evaluationsOfficial $10 / $5095.2
05GPT-5.6 SolOpenAIEvidence-sensitive · 5 evaluationsOfficial $4 / $2093.9
06claude-sonnet-5.5anthropicEvidence-sensitive · 5 evaluationsOfficial —92.5
07GPT-6 SolOpenAIEvidence-sensitive · 5 evaluationsOfficial $2 / $1091.2
08muse-wonder-1.0—Evidence-sensitive · 4 evaluationsOfficial —90.7
09Claude Opus 5AnthropicEvidence-sensitive · 5 evaluationsOfficial $5 / $2590.7
10Claude Fable 5AnthropicEvidence-sensitive · 5 evaluationsOfficial $10 / $5090.2
11GPT-5.6 TerraOpenAIEvidence-sensitive · 5 evaluationsOfficial $2 / $1287.1
12GPT-5.5OpenAIEvidence-sensitive · 5 evaluationsOfficial $5 / $3086.3
13Claude Opus 4.8AnthropicEvidence-sensitive · 5 evaluationsOfficial $5 / $2581.1
14Muse Spark 1.3MetaEvidence-sensitive · 5 evaluationsOfficial $1.25 / $4.2579.8
15Gemini 3.7 FlashGoogleEvidence-sensitive · 5 evaluationsOfficial $0.75 / $3.7577.8
16Grok 4.6xAIEvidence-sensitive · 5 evaluationsOfficial $2 / $677.0
17Claude Sonnet 5AnthropicEvidence-sensitive · 5 evaluationsOfficial $2 / $1075.4
18Muse Spark 1.2MetaEvidence-sensitive · 2 evaluationsOfficial $1.25 / $4.2572.4
19DeepSeek V4 Pro 0813DeepSeekEvidence-sensitive · 5 evaluationsOfficial $1.35 / $4.0471.1
20Gemini 3.8 FlashGoogleEvidence-sensitive · 5 evaluationsOfficial $0.75 / $3.7571.0
21Qwen3.8 MaxAlibabaEvidence-sensitive · 5 evaluationsOfficial $1.79 / $5.3870.2
22GPT-5.6 LunaOpenAIEvidence-sensitive · 5 evaluationsOfficial $0.2 / $1.269.3
23claude-haiku-5-5—Evidence-sensitive · 5 evaluationsOfficial —67.5
24Claude Opus 4.7AnthropicEvidence-sensitive · 5 evaluationsOfficial $5 / $2567.0
25DeepSeek V4.1 FlashDeepSeekEvidence-sensitive · 5 evaluationsOfficial $0.3 / $1.265.7
26GPT-5.2 2025-12-11OpenAIEvidence-sensitive · 5 evaluationsOfficial $1.75 / $1462.9
27Grok 4.7xAIEvidence-sensitive · 5 evaluationsOfficial $2 / $662.5
28Kimi K3Moonshot AIEvidence-sensitive · 5 evaluationsOfficial $2.99 / $14.9662.0
29Grok 4.5xAIEvidence-sensitive · 4 evaluationsOfficial $2 / $661.7
30Claude Opus 4.6AnthropicEvidence-sensitive · 5 evaluationsOfficial $5 / $2560.5

Each board shows at most 30 models

The consensus index is not an accuracy rate; ties keep the order the shared evidence establishes.

How to read this board

Aggregates several public evaluations; models differ in which ones they take part in.

See how it is computed →

About prices

API prices are shown in USD per million tokens by default. Vendor list prices and the DogRouter reference sit in separate columns; the latter is the relay price of a Dogpay sister service, carries a verification date and a link to its price page, is never shown as a discount, and does not affect the ranking. Cached price is the input rate on a cache hit; cache writes, storage and subscriptions are billed separately. CNY list prices are converted at the 2026-10-16 rate (欧洲央行).