Skip to content
OpenSmartRoute
Model catalogue
T

Thinking Machines

Inkling (batch)

thinkingmachines/inkling:batch

Specification

Published by the vendor; the platform bills at the target's declared price.

T

Inkling (batch)

reasoning open weights tools

Thinking Machines thinkingmachines/inkling:batch

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,...

Input $/1M
$1.00
Output $/1M
$4.05
Context
524K
Max output
472K
Released
Jul 17, 2026
Modalities
textimageaudio
Intelligence index32
Coding index52
Agentic index24

Cached input reads $0.17 per 1M · weights on Hugging Face as thinkingmachines/Inkling

On this deployment

No routing target resolves to this model yet. Declare one and the router picks it up without retraining.

Not routable here yet
Add a target with metadata.model set to this id.
targets.yamlyaml
# targets.yaml - declare a target that runs Inkling (batch)
targets:
  - id: inkling-batch
    kind: llm
    name: Inkling (batch)
    metadata:
      model: thinkingmachines/inkling:batch          # links the target to this catalogue entry
    capabilities:
      domains: [general]
      supports_tools: true
      context_window: 524288
    cost:
      usd_per_1k_tokens: 0.002525

Price among Thinking Machines models

Input list price per million tokens; Inkling (batch) is the 5th cheapest of 6 listed.