Skip to content

Model vendor

Thinking Machines models: API pricing and providers

1 Thinking Machines model with Thinking Machines's published list rates and the providers that serve each one. You call them through INFRO on a provider account you connect; that account bills you, and INFRO adds 0% to model spend.

Rates checked 2026-10-05.

Thinking Machines models

ModelModalityList rateProviders
Inkling
thinkingmachines/inkling
Open-weight multimodal generalist (975B MoE)
Text$1.00 per 1M input tokens and $4.05 per 1M output tokens, with cached input at $0.17 per 1M tokens3

Thinking Machines models: questions

Which Thinking Machines models does this reference list?
1 Thinking Machines model: 1 text. Each has its own page with Thinking Machines's list rate and the providers that serve it.
Which providers serve Thinking Machines models?
Together AI, Fireworks AI and OpenRouter. You connect your own account with any of them; INFRO routes between the ones you connect.
What is the lowest Thinking Machines input rate listed?
Inkling, at $1.00 per 1M input tokens and $4.05 per 1M output tokens, with cached input at $0.17 per 1M tokens on Thinking Machines's list (checked 2026-10-05).
Does INFRO add a markup to Thinking Machines model usage?
No. Your Thinking Machines or provider account bills usage directly; INFRO charges a software subscription and takes 0% of model spend.

Other model vendors

Search and filter every model