Robots Atlas>ROBOTS ATLAS
Muse Spark 1.1
AI Models›Muse

Muse Spark 1.1

Family: Muse
Muse Spark 1.1 is an update to Meta's (Meta Superintelligence Labs) natively multimodal Muse-family model. It introduces a 1M-token context and major gains in tool use, computer use, coding and multimodal understanding; the public preview of the Meta Model API launched alongside it.
✓ ActiveLLMMultimodalReasoning modelTool-using model📁 Muse
Context window
1 mln tokenow
tokens
Release date
9 July 2026
🏢Meta AIProducer
Access:APIHostedDeployment:☁ Cloud

Overview

Muse Spark 1.1 is an update to Muse Spark - the natively multimodal reasoning model developed by Meta Superintelligence Labs (MSL), the first in the Muse family. Version 1.1 was released on July 9, 2026.

Its key new feature is a 1-million-token context, actively managed by the model. Version 1.1 brings major gains in tool use, computer use, coding and multimodal understanding, advancing the performance-efficiency frontier. It also improves adversarial robustness and reduces hallucination rates and sycophancy.

Alongside Muse Spark 1.1, Meta launched the public preview of the Meta Model API, and the model is available in 'Thinking' mode in the Meta AI app and on meta.ai. It is praised for combining a million-token context with strong agentic capabilities, computer use and visual-to-code artifact generation; commentators on the release included Amjad Masad, CEO of Replit.

Classification
LLMMultimodalReasoning modelTool-using model
Family: Muse
Access & deployment
APIHosted
Cloud
Weights: Closed
Key parameters
📏 Context: 1 mln tokenow
✓ Tools
📥 Input: text, image

Technical specification

Context window
1 mln tokenow
tokens
License
Proprietary
Features:✓ Tool use
Modalities
⬇ Input
textimage
⬆ Output
text

Capabilities and applications

Native model capabilities
Reasoning
The model's ability to reason logically and solve complex problems.
Category: reasoning
Multimodal understanding
Category: multimodal
Image understanding
Analysing and interpreting the content of images.
Category: vision
Chart understanding
Reading and interpreting charts, tables and diagrams.
Category: vision
Coding
Generating, analysing and modifying code in many programming languages. Covers writing functions, debugging, refactoring, code review, and creating tests. Measured by benchmarks such as HumanEval and SWE-bench.
Category: coding

Technical architecture

Core Architecture
COCoTMAMAS
Model Form
MLMultimodal LLMRMReasoning model
Training Techniques
RLRLHF

Sources and related pages