thinkingmachines Open weights Vision Reasoning

Inkling pricing & benchmarks

Released Jul 17, 2026 · served by 3 providers · ranked #27 by intelligence in our index

Input / 1M tokens$1.00
Output / 1M tokens$4.05
Context window1.0M
Cached input$0.170
Intelligence index40.7
Coding index52.1
Agentic index32.3
1M in + 300K out$2.21

What Inkling is

Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...

Providers and prices for Inkling

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
DeepInfracheapest $1.00 $4.05 524K fp8 96.4%
BaseTen $1.00 $4.05 1.0M fp8 99.9%
Together $1.00 $4.05 524K 99.2%

Cheaper models in the same class

Models scoring within 4 points of Inkling on the Intelligence Index, but with a lower output price.

FAQ

How much does the Inkling API cost?

$1.00 per 1M input tokens and $4.05 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $2.21.

Is Inkling free?

No. It is a paid model starting at $1.00 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for Inkling?

DeepInfra at $1.00 input / $4.05 output per 1M tokens (fp8 quantization).

Can I self-host Inkling?

Yes — weights are published as thinkingmachines/Inkling on Hugging Face, so you can run it on your own hardware.