Open Source   Thinking Machines Lab · Released July 15, 2026

Inkling

Mira Murati's Lab Launches a 975B Open Multimodal Model

Architecture Mixture-of-Experts transformer with speculative MTP layers
Parameters 975B total — 41B active parameters (12B active for Inkling-Small)
Context Window 1 million tokens
Modalities Text, Images, Audio — native any-to-any multimodal
Input Pricing 50% launch discount on Tinker (time-limited)
Output Pricing Self-hostable via open weights — no API cost
Open Weights Yes — full BF16 and NVFP4 checkpoints on Hugging Face
Availability Hugging Face (download), Tinker (hosted), SGLang/vLLM/llama.cpp (inference day-0)
Why This Model Matters Inkling is Thinking Machines Lab's first general-purpose AI model and the most significant product from the company founded by former OpenAI CTO Mira Murati. It is a 975-billion-parameter Mixture-of-Experts system trained from scratch on 45 trillion tokens of text, images, audio, and video — making it one of the largest openly available multimodal models in the world. The model natively processes text, images, and audio in any combination without separate adapters, and includes speculative Multi-Token Prediction (MTP) layers for faster inference. What makes Inkling particularly notable is its accessibility: day-0 support in Hugging Face transformers (v5.14.0), SGLang, vLLM, and llama.cpp means developers can start using it immediately on existing infrastructure. Alongside the main model, Thinking Machines previewed Inkling-Small — a 276B-parameter MoE with 12B active parameters that matches or exceeds its larger sibling on many benchmarks through improved training data and recipe. The model is available in full BF16 precision for Hopper-class GPUs and quantized NVFP4 format optimized for NVIDIA Blackwell systems.
Visit Thinking Machines Lab → ← All Releases
Performance

Benchmarks & Evaluation Results

Reasoning

Matches or exceeds larger MoE models on multi-step reasoning tasks

Agentic Tasks

Strong performance on tool-use and autonomous coding benchmarks

Multimodal

Competitive with frontier models on vision-language evaluations

Inkling-Small

Matches Inkling on many benchmarks despite 3.5x fewer active params

In Depth

Full Analysis of Inkling

Inkling represents Thinking Machines Lab's vision for open, customizable AI. Founded in February 2025, the company has been deliberately deliberate about its first model release — publishing novel research, building a customization platform, and previewing collaborative AI systems before shipping Inkling. The model itself is designed as a broad, balanced foundation model rather than a narrow benchmark optimizer. It was trained on an unusually diverse dataset covering text, images, audio, and v

ideo across 45 trillion tokens, with the goal of creating a flexible base that can be fine-tuned for specialized applications. The MoE architecture activates only 41 billion of its 975 billion parameters per token, making inference practical on clusters of 4-8 GPUs. The speculative MTP layers predict multiple future tokens simultaneously, improving generation speed without sacrificing quality. Thinking Machines has made Inkling fully available on Hugging Face under an open weights model, with bo

th BF16 and NVFP4 quantized formats. The company is also offering hosted inference on its Tinker platform at a 50% launch discount. Inkling-Small, the 12B-active-parameter variant, is particularly interesting: because of improvements in training data quality and recipe, it matches Inkling on many evaluations while being dramatically cheaper to run.

Assessment

Strengths & Considerations

Strengths

  • + Native any-to-any multimodal (text + image + audio) without separate adapters
  • + Fully open weights on Hugging Face — download, fine-tune, self-host
  • + Day-0 support in all major inference engines (transformers, SGLang, vLLM, llama.cpp)
  • + Speculative MTP layers for faster inference throughput
  • + Two size options for different cost targets (41B and 12B active)
  • + NVFP4 quantization for efficient Blackwell GPU inference
  • + Backed by Mira Murati's team with deep transformer expertise

Considerations

  • Trails top proprietary models on some frontier benchmarks
  • Inkling-Small still in preview; full release pending
  • 50% launch discount is time-limited
  • Smaller ecosystem and community compared to Llama or Qwen families
  • 975B total parameters still requires multi-GPU setup for self-hosting
Applications

Best Use Cases for Inkling

Where this model excels and the types of workloads it is best suited for.

Use Case 1

Multimodal applications needing native image, audio, and text understanding

Use Case 2

Self-hosted deployments for data-sensitive workloads

Use Case 3

Fine-tuning and customization for specialized domains

Use Case 4

Research requiring full model access and modification

Use Case 5

Synthetic data generation across multiple modalities

Use Case 6

Applications needing controllable thinking effort vs. cost tradeoffs

Getting Started

How to Use Inkling

Quick Start Guide

Download Inkling from Hugging Face at huggingface.co/thinkingmachines/Inkling. Install transformers v5.14.0+ (pip install -U transformers) and use the any-to-any pipeline for multimodal inference. For hosted access, visit Tinker at thinkingmachines.ai — currently at 50% launch discount. For self-hosting, the BF16 checkpoint requires Hopper-class GPUs; the NVFP4 checkpoint is optimized for Blackwell. SGLang and vLLM both support Inkling with day-0 integration.

Visit Thinking Machines Lab → ← All July 2026 Releases