The Orchestrator › Models › Thinking Machines

Thinking Machines: Inkling 2 Small

Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is the first model in the second...

Specifications

ProviderThinking Machines
Model IDthinkingmachines/inkling-2-small
Context window1.0M tokens (1,048,576)
Modalitiestext + image → text
TokenizerOther
Router sourcesOpenRouter, TrustedRouter, Vercel AI Gateway, Cloudflare
Released2026-11-09
Tracked since2026-11-14
PopularityNot in the ranked set
ReasoningMandatory

Pricing (per million tokens)

VariantSourceAPI IDInputOutputContext
standardOpenRouterthinkingmachines/inkling-2-small$0.3$1.001.0M
freeOpenRouterthinkingmachines/inkling-2-small:freeFreeFree1.0M
standardTrustedRouterthinkingmachines/inkling-2-small$0.317$1.061.0M
standardVercel AI Gatewaythinkingmachines/inkling-2-small$0.3$1.001.0M
standardCloudflarethinkingmachines/inkling-2-small$0.3$1.00—

Recorded price history

DateInput / 1MOutput / 1MContext
2026-11-14$0.3$1.00—
2026-11-14$0.3$1.001.0M
2026-11-14$0.317$1.061.0M
2026-11-14$0.3$1.001.0M

Capabilities

Vision (image input)Agentic tool useStructured outputsReasoning (always on)1M+ context windowControllable reasoningNo content storage (attested)Zero data retentionNo training on inputs (attested)

Trust flags come from TrustedRouter's per-provider attestations.

Supported parameters

frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p