Thinking Machines: Inkling

Image input Audio input Text input Text output
Author's Description

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,...

Key Specifications
Cost
$$$$$
Context
524K
Parameters
975B (Rumoured)
Released
Jul 17, 2026
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Include Reasoning Max Tokens Logit Bias Tools Tool Choice Reasoning Top P Min P Frequency Penalty Presence Penalty Temperature
Features

This model supports the following features:

Tools Reasoning
Performance Summary

Thinking Machines' Inkling, an open-weight multimodal mixture-of-experts model with 41B active parameters, demonstrates a balanced performance profile. It exhibits moderate speed, ranking in the 26th percentile, and offers moderate pricing, placing in the 21st percentile across benchmarks. A standout feature is its exceptional reliability, boasting a 99% success rate, indicating minimal technical failures. Inkling excels in several key areas. It achieves perfect accuracy in General Knowledge and Ethics, with the former being the most accurate model at its price point and speed. Its Reasoning capabilities are strong, scoring 97.9% accuracy (83rd percentile), and it performs very well in Mathematics (89th percentile). Instruction Following and Coding also show solid performance, ranking in the 75th and 72nd percentiles respectively. While its Hallucinations score of 95.8% is respectable, it indicates some room for improvement in acknowledging uncertainty compared to models that achieve 100% in this specific test. Overall, Inkling presents itself as a highly reliable and capable model, particularly strong in knowledge, reasoning, and ethical considerations, making it well-suited for general-purpose reasoning, coding, and agentic systems.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1
Completion $4.05
Input Cache Read $0.17

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Together
Together | thinkingmachines/inkling-20260715 524K $1 / 1M tokens $4.05 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration