AimodraclaudeMost capable
claude-fable-5
claude-fable-5$4.20 input · $6.60 outputper 1M tokens
AI model pricing comparison
for 47 models.
Compare AI model prices, exact IDs and live availability across Claude, GPT, Gemini, Grok, Kimi, GLM, Qwen and DeepSeek before you integrate.
Open the API integration guideChecking model availability…
Loading current models…
Only available IDs are included in bulk copyHow live AI model availability works
Available means at least one configured route passed a recent text check. Currently down means all enabled routes failed their latest checks. Failed models are checked approximately every 15 minutes; healthy models approximately hourly. Unchecked or stale results show Not verified. These checks do not guarantee every request or confirm image support.
claudeMost capable
claude-fable-5-1
claude-fable-5-1$4.20 input · $6.60 outputper 1M tokens
claudeMost capable
claude-opus-4-7
claude-opus-4-7$2.60 input · $4.20 outputper 1M tokens
claudeMost capable
claude-opus-4-8
claude-opus-4-8$2.60 input · $4.20 outputper 1M tokens
claudeMost capable
claude-opus-5
claude-opus-5$2.60 input · $5.00 outputper 1M tokens
gptFastest
gpt-5.6-luna
gpt-5.6-luna$0.20 input · $0.40 outputper 1M tokens
gptMost capable
gpt-5.6-sol
gpt-5.6-sol$3.40 input · $5.80 outputper 1M tokens
gptMost capable
gpt-6-astra
gpt-6-astra$5.05 input · $9.10 outputper 1M tokens
AI model pricing and access FAQ
Is there a cheaper way to access multiple AI models?
Aimodra publishes every model's input and output rate per million tokens and charges usage against a fixed monthly allowance. Compare prices and live availability before you integrate.
Can I use Claude and GPT with one API key?
Yes. One Aimodra API key calls 47 configured model IDs across Claude, GPT, Gemini, Grok, Kimi, GLM, Qwen and DeepSeek. Isolate keys by project and revoke access anytime.
Where can I compare AI model prices?
This directory lists each model ID, family, input rate and output rate in US dollars per million tokens. Live availability appears beside every model.
How often is model availability checked?
Failed models are checked approximately every 15 minutes and healthy models approximately hourly. Unchecked or stale routes show Not verified.