gpt-5.4
Input$1.7000/ 1M tokens
Output$10.2000/ 1M tokens
Model Pricing and Catalog
Compare current API prices, access groups, and suitable models for Claude Code, Codex, and other tools.
Types
Providers
Group
Pricing currency
gpt-5.4-mini
Input$0.5100/ 1M tokens
Output$3.0600/ 1M tokens
gpt-5.5
Input$3.4000/ 1M tokens
Output$20.4000/ 1M tokens
gpt-5.6-luna
Input$0.1360/ 1M tokens
Output$0.8160/ 1M tokens
gpt-5.6-sol
Input$3.4000/ 1M tokens
Output$20.4000/ 1M tokens
gpt-5.6-terra
Input$1.3600/ 1M tokens
Output$8.1600/ 1M tokens
gpt-image-2
Input$0.200/ request
OutputN/A
gpt-image-2-all
Input$0.100/ request
OutputN/A
claude-fable-5
Input$6.8000/ 1M tokens
Output$34.0000/ 1M tokens
claude-haiku-4-5-20251001
Input$0.6800/ 1M tokens
Output$3.4000/ 1M tokens
claude-opus-4-6
Input$3.4000/ 1M tokens
Output$17.0000/ 1M tokens
claude-opus-4-7
Input$3.4000/ 1M tokens
Output$17.0000/ 1M tokens
claude-opus-4-8
Input$3.4000/ 1M tokens
Output$17.0000/ 1M tokens
claude-opus-5
Input$3.4000/ 1M tokens
Output$17.0000/ 1M tokens
claude-sonnet-4-6
Input$2.0400/ 1M tokens
Output$10.2000/ 1M tokens
claude-sonnet-5
Input$1.3600/ 1M tokens
Output$6.8000/ 1M tokens
deepseek-v4-flash
Input$0.1080/ 1M tokens
Output$0.2170/ 1M tokens
deepseek-v4-flash-0731
Input$0.0950/ 1M tokens
Output$0.1910/ 1M tokens
deepseek-v4-pro
Input$0.3370/ 1M tokens
Output$0.6740/ 1M tokens
glm-5.2
Input$1.2320/ 1M tokens
Output$3.8720/ 1M tokens
kimi-k3
Input$2.7600/ 1M tokens
Output$13.8000/ 1M tokens
MiniMax-M3
Input$0.4080/ 1M tokens
Output$1.6320/ 1M tokens
qwen3.8-max
Input$1.8400/ 1M tokens
Output$5.5200/ 1M tokens
Live API catalog
How to read BetterToken pricing
The catalog uses current service data: available models, currency, access groups, and ratios come from the pricing API, not from a static price sheet.
Supported models
The catalog currently includes 23 models from 7 providers. Available types: Text, Images.
How billing works
Text models show input and output token prices separately. When a model supports cache, image, or audio billing, those rows appear in the model card details and group pricing.
How to compare models
Compare more than price: check provider, endpoint type, access group, model description, and output price, because long answers usually drive the final cost.
Estimate Claude Code and Codex
Estimate the input context and expected output, then multiply them by the selected model input and output prices. For Codex, review GPT group models; for Claude Code, review Claude group models.
Pricing questions
Prices come from the current BetterToken pricing API and update with model, currency, and access group settings.
Choose a suitable GPT model, estimate task input and output tokens, then calculate the cost from that model input and output prices.
Open Claude models, choose an available group, and use input and output token prices to estimate context, files, and expected response size.