Skip to content
PAY.NEWS LEADERBOARD

AI model leaderboard

Evaluation sources How ranks are computed

Real evaluations across many capabilities, to find the models that perform best overall.

19 evaluations·8 operators·10/17 20:05 updated

Overall 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
01gemini-4-argonGoogleEvidence-sensitive · 11 evaluationsOfficial —86.7
02Claude Opus 5.5AnthropicEvidence-sensitive · 18 evaluationsOfficial $4 / $2085.9
03GPT-6 AstraOpenAIEvidence-sensitive · 19 evaluationsOfficial $10 / $5085.5
04Claude Fable 5.1AnthropicEvidence-sensitive · 17 evaluationsOfficial $10 / $5085.1
05Claude Fable 5AnthropicEvidence-sensitive · 18 evaluationsOfficial $10 / $5084.0
06Gemini 3.7 FlashGoogleEvidence-sensitive · 17 evaluationsOfficial $0.75 / $3.7580.0
07gpt-6.1-sol—Evidence-sensitive · 16 evaluationsOfficial —79.4
08Claude Opus 5AnthropicEvidence-sensitive · 18 evaluationsOfficial $5 / $2578.2
09Gemini 3.8 FlashGoogleEvidence-sensitive · 19 evaluationsOfficial $0.75 / $3.7578.0
10GPT-5.6 SolOpenAIEvidence-sensitive · 19 evaluationsOfficial $4 / $2077.5
11claude-sonnet-5.5anthropicEvidence-sensitive · 16 evaluationsOfficial —76.5
12Muse Spark 1.3MetaEvidence-sensitive · 15 evaluationsOfficial $1.25 / $4.2575.5
13GPT-5.5OpenAIEvidence-sensitive · 18 evaluationsOfficial $5 / $3071.8
14Qwen3.8 MaxAlibabaEvidence-sensitive · 16 evaluationsOfficial $1.79 / $5.3871.5
15GPT-6 SolOpenAIEvidence-sensitive · 18 evaluationsOfficial $2 / $1070.9
16Claude Opus 4.7AnthropicEvidence-sensitive · 17 evaluationsOfficial $5 / $2568.0
17Claude Opus 4.8AnthropicEvidence-sensitive · 18 evaluationsOfficial $5 / $2566.5
18Kimi K3Moonshot AIEvidence-sensitive · 18 evaluationsOfficial $2.99 / $14.9665.6
19Qwen3.8 Max 0902AlibabaEvidence-sensitive · 7 evaluationsOfficial $1.79 / $5.3865.3
20GLM-5.3Z.aiEvidence-sensitive · 18 evaluationsOfficial $1.4 / $4.464.2
21Muse Spark 1.2MetaEvidence-sensitive · 14 evaluationsOfficial $1.25 / $4.2563.5
22Claude Opus 4.6AnthropicEvidence-sensitive · 16 evaluationsOfficial $5 / $2563.2
23Muse Spark 1.1MetaEvidence-sensitive · 14 evaluationsOfficial $1.25 / $4.2559.5
24Gemini 3 Pro PreviewGoogleEvidence-sensitive · 8 evaluationsOfficial —58.9
25Gemini 3.6 FlashGoogleEvidence-sensitive · 17 evaluationsOfficial $0.75 / $3.7558.0
26GPT-5.6 TerraOpenAIEvidence-sensitive · 19 evaluationsOfficial $2 / $1257.3
27Grok 4.6xAIEvidence-sensitive · 16 evaluationsOfficial $2 / $656.8
28GPT-5.4OpenAIEvidence-sensitive · 11 evaluationsOfficial $2.5 / $1556.6
29DeepSeek V4.1 FlashDeepSeekEvidence-sensitive · 17 evaluationsOfficial $0.3 / $1.254.1
30claude-haiku-5-5—Evidence-sensitive · 12 evaluationsOfficial —52.2

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).