Back to News Hub
🤗Hugging Face
July 15, 2026
General AI

Welcome Inkling by Thinking Machines

Overview

We're on a journey to advance and democratize artificial intelligence through open source and open science. Back to Articles a]:hidden"> Welcome Inkling by Thinking Machines Published July 15, 2026 Update on GitHub Upvote 8 +2 ben burtenshaw burtenshaw Follow merve merve Follow Pedro Cuenca pcuenq Follow Aritra Roy Gosthipaty ariG23498 Follow Inkling is a large (1T params! ) open model to natively accept image, text, and audio inputs.

Key Takeaways

  • TLDR; Inkling by Thinking Machines is out on Hugging Face.

    Inkling is a huge multimodal LLM that understands all modalities (image, audio, text), has agentic capabilities, and supports 1M context.

  • Overall Capabilities and Architecture Inkling is a decoder-only multimodal Mixture-of-Experts model with 975B total and 41B active parameters.

    There are a lot of things going on, so let's break each part down: Decoder-only: This means that the architecture supports causal autoregressive generation, like in most state-of-the-art LLMs.

  • Each attention layer learns position directly in the attention logits.

    Aside from key-query-values, there's a fourth projection producing a per-token, per-head relative feature R.

  • Short convolution: The model uses a distinctive short 1D convolution, or over the hidden states.

    SConv reads the current token and the previous hidden states, with being the sliding window size.

  • The multimodal towers are relatively simple modules, unlike other models that employ separate encoders for each modality.

Stats & Key Facts

  • #Inkling is the first large open model with ~1T parameters and 1M context window to natively receive image, text, and audio inputs , trained on 45 trillion tokens of text, images, audio and video.
  • #The model has 256 experts, as we'll see later.

TLDR; Inkling by Thinking Machines is out on Hugging Face. Inkling is a huge multimodal LLM that understands all modalities (image, audio, text), has agentic capabilities, and supports 1M context. It comes in full BF16 and a well-calibrated NVFP4 variant, and includes speculative MTP layers for faster inference.

There's day-0 support in transformers, SGLang, and llama. Inkling is the first large open model with ~1T parameters and 1M context window to natively receive image, text, and audio inputs , trained on 45 trillion tokens of text, images, audio and video. It's focused on reasoning across modalities such as audio, images, and text; and is intended for domain adaptation via fine-tuning.

We've tinkered with this model to build some demos and explore the architecture, and we think it's great for building a new wave of multimodal reasoning apps. Overall Capabilities and Architecture Inkling is a decoder-only multimodal Mixture-of-Experts model with 975B total and 41B active parameters. There are a lot of things going on, so let's break each part down: Decoder-only: This means that the architecture supports causal autoregressive generation, like in most state-of-the-art LLMs.

For more details please read the original article at Hugging Face.

Continue Learning

Originally published by Hugging Face
Read the original

Comments

Sign in to join the conversation