Skip to content
PAY.NEWS LEADERBOARD

AI model leaderboard

Evaluation sources How ranks are computed

From writing code to reshaping a repository — can the model actually ship software?

3 evaluations·3 operators·10/17 20:05 updated

Coding 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
01Claude Opus 5.5AnthropicEvidence-sensitive · 2 evaluationsOfficial $4 / $2096.0
02Claude Fable 5.1AnthropicEvidence-sensitive · 2 evaluationsOfficial $10 / $5095.1
03claude-sonnet-5.5anthropicEvidence-sensitive · 2 evaluationsOfficial —92.0
04Claude Opus 5AnthropicEvidence-sensitive · 2 evaluationsOfficial $5 / $2591.2
05GPT-6 AstraOpenAIEvidence-sensitive · 3 evaluationsOfficial $10 / $5090.0
06GPT-5.6 SolOpenAIEvidence-sensitive · 3 evaluationsOfficial $4 / $2088.4
07Gemini 3.8 FlashGoogleEvidence-sensitive · 3 evaluationsOfficial $0.75 / $3.7588.0
08DeepSeek V4.1 FlashDeepSeekEvidence-sensitive · 2 evaluationsOfficial $0.3 / $1.287.1
09Claude Fable 5AnthropicEvidence-sensitive · 2 evaluationsOfficial $10 / $5085.0
10GLM-5.3Z.aiEvidence-sensitive · 3 evaluationsOfficial $1.4 / $4.482.1
11Kimi K3Moonshot AIEvidence-sensitive · 3 evaluationsOfficial $2.99 / $14.9680.6
12Gemini 3.7 FlashGoogleEvidence-sensitive · 2 evaluationsOfficial $0.75 / $3.7575.5
13GPT-5.5OpenAIEvidence-sensitive · 2 evaluationsOfficial $5 / $3074.6
14GLM-5.3 FlashZ.aiEvidence-sensitive · 3 evaluationsOfficial $0.12 / $0.41970.1
15Qwen3.8 MaxAlibabaEvidence-sensitive · 2 evaluationsOfficial $1.79 / $5.3869.1
16Claude Sonnet 5AnthropicEvidence-sensitive · 2 evaluationsOfficial $2 / $1068.5
17Muse Spark 1.1MetaEvidence-sensitive · 2 evaluationsOfficial $1.25 / $4.2565.8
18Muse Spark 1.2MetaEvidence-sensitive · 2 evaluationsOfficial $1.25 / $4.2565.6
19gpt-6.1-sol—Evidence-sensitive · 2 evaluationsOfficial —63.4
20GPT-6 SolOpenAIEvidence-sensitive · 2 evaluationsOfficial $2 / $1058.1
21claude-haiku-5-5—Evidence-sensitive · 2 evaluationsOfficial —57.6
22GPT-5.6 TerraOpenAIEvidence-sensitive · 3 evaluationsOfficial $2 / $1252.1
23Grok 4.6xAIEvidence-sensitive · 2 evaluationsOfficial $2 / $651.2
24Claude Opus 4.8AnthropicEvidence-sensitive · 2 evaluationsOfficial $5 / $2544.6
25GPT-6 LunaOpenAIEvidence-sensitive · 2 evaluationsOfficial $0.1 / $0.543.6
26GPT-5.6 LunaOpenAIEvidence-sensitive · 3 evaluationsOfficial $0.2 / $1.243.6
27DeepSeek V4 Pro 0813DeepSeekEvidence-sensitive · 2 evaluationsOfficial $1.35 / $4.0442.1
28DeepSeek V4 Pro PreviewDeepSeekEvidence-sensitive · 2 evaluationsOfficial $0.435 / $0.8739.1
29Grok 4.5xAIEvidence-sensitive · 2 evaluationsOfficial $2 / $637.6
30GPT-5.4OpenAIEvidence-sensitive · 2 evaluationsOfficial $2.5 / $1536.8

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).