thinkingmachines

Thinking Machines: Inkling Small

Name: Thinking Machines: Inkling Small
Creator: thinkingmachines
Published: 2026-07-30T20:25:17.000Z

Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...

Vedi la fonte su OpenRouter Hugging Face

Panoramica

Specifiche del modello

Context: 1.048.576 tokens
Output massimo: 262.144 tokens
Architettura: text+image+audio->text
Tokenizer: Other
Limite di conoscenza: Not provided
Moderato: No

Capacità e modalità

Input → Output

Input

textimageaudio

Output

text

Ragionamento

Sì · max · high · medium · low · minimal · none

API

Parametri API supportati

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyseedstoptemperaturetool_choicetoolstop_ktop_logprobstop_p

Provider disponibili

3 provider

DeepInfra

fp8

Context: 524K
Output massimo: 262K
Input: $0.450
Output: $1.20
Lettura cache: $0.100
Scrittura cache: Not provided

BaseTen

fp8

Context: 1.0M
Output massimo: 33K
Input: $0.500
Output: $1.20
Lettura cache: $0.100
Scrittura cache: Not provided

Together

unknown

Context: 524K
Output massimo: 472K
Input: $0.500
Output: $1.20
Lettura cache: $0.100
Scrittura cache: Not provided

thinkingmachines

models

Tutti i modelli

thinkingmachines

Pubblicato 31 lug 2026