Average Ratings 1 Rating

Total 5 out of 5 stars
ease 5 out of 5 stars

Average Ratings 1 Rating

Total 5 out of 5 stars

Description

Muse Code is a beta terminal coding agent from Meta designed to help developers complete complex software engineering work across large codebases. Powered by Muse Spark 1.2, the agent can plan repository changes, write code, run validation steps, and coordinate persistent subagents for difficult development tasks. Muse Code uses a simple main agent loop supported by async background agents that remain active throughout each session. These background agents can gather information, carry out next steps, and decide when to report back to the main agent, reducing latency and unnecessary user steering. The runtime is built around a local event log where every model call, tool run, approval, and edit is appended. This event log makes Muse Code replay-exact and restart-safe, allowing it to resume from the point of failure after a crash. Muse Code also ships with default skills, including /plan, /grill, and /goal, to support structured planning, plan validation, and objective completion. It can be installed on macOS or Linux and is integrated with Meta’s AI developer ecosystem. By combining terminal-based coding, persistent subagents, replay-safe execution, bundled skills, and Muse Spark 1.2, Muse Code helps developers automate larger and longer software engineering workflows.

Description

Muse Glimmer is an open-weights model featuring 30 billion parameters, developed by Meta Superintelligence Labs, and is fine-tuned for continuous local agent operations. Its compact design allows it to function on a standard Mac or PC equipped with a single consumer GPU, making it ideal for various tasks such as local agent management, function calling, programming, and LLM-as-a-judge evaluations without reliance on cloud services or internet connectivity. This innovative model integrates advanced capabilities such as long-horizon execution, accurate tool invocation, multimodal comprehension, extended memory for context, and effective instruction following. It is proficient in accomplishing end-to-end tasks as an agent, maintains the ability to engage in multi-step reasoning over lengthy processes, can recover gracefully from failed or unanticipated tool engagements, and interprets interleaved text and images using a specialized perception encoder designed for analyzing screenshots, graphs, and document files. Furthermore, Muse Glimmer is compatible with OpenClaw and other orchestration frameworks, allowing for adjustable reasoning efforts, and has been developed with a diverse dataset encompassing over 100 languages. The model's versatility ensures that it can adapt to various applications, thus enhancing its utility in different domains.

API Access

Has API Yes 

API Access

Has API Yes 

Screenshots View All

Screenshots View All

Integrations

Meta AI Yes 
CodeAgentSwarm Yes 
CueHQ Yes 
ExecuTorch No 
Hermes Agent No 
Hugging Face No 
LM Studio No 
Muse Code No 
Muse Ember Yes 
Muse Glimmer Yes 
Muse Spark Yes 
Muse Spark 1.1 Yes 
Muse Spark 1.2 Yes 
Muse Spark 1.3 Yes 
Muse Wonder Yes 
Muse Wonder Full Yes 
Ollama No 
OpenClaw No 
OpenCode No 
Unsloth No 

Integrations

Meta AI Yes 
CodeAgentSwarm No 
CueHQ No 
ExecuTorch Yes 
Hermes Agent Yes 
Hugging Face Yes 
LM Studio Yes 
Muse Code Yes 
Muse Ember No 
Muse Glimmer No 
Muse Spark No 
Muse Spark 1.1 No 
Muse Spark 1.2 No 
Muse Spark 1.3 No 
Muse Wonder No 
Muse Wonder Full No 
Ollama Yes 
OpenClaw Yes 
OpenCode Yes 
Unsloth Yes 

Pricing Details

$1.25 per 1M tokens (input)
Standard pricing: $1.25 per million input tokens and $4.25 per million output tokens

Discounted "contributor" tier costing $0.10 per million input tokens and $0.20 per million output tokens for users who agree to share feedback to improve the AI.
Free Trial No 
Free Version Yes 

Pricing Details

Free
Free Trial Yes 
Free Version No 

Deployment

Web-Based No 
On-Premises No 
iPhone App No 
iPad App No 
Android App No 
Windows Yes 
Mac Yes 
Linux Yes 
Chromebook No 

Deployment

Web-Based Yes 
On-Premises No 
iPhone App No 
iPad App No 
Android App No 
Windows No 
Mac No 
Linux No 
Chromebook No 

Customer Support

Business Hours No 
Live Rep (24/7) No 
Online Support Yes 

Customer Support

Business Hours No 
Live Rep (24/7) No 
Online Support Yes 

Types of Training

Training Docs Yes 
Webinars No 
Live Training (Online) No 
In Person No 

Types of Training

Training Docs Yes 
Webinars No 
Live Training (Online) No 
In Person No 

Vendor Details

Company Name

Meta

Founded

2004

Country

United States

Website

meta.ai

Vendor Details

Company Name

Meta

Founded

2004

Country

United States

Website

meta.ai/

Alternatives

GitHub Copilot Reviews

GitHub Copilot

GitHub
4.0 out of 5 stars

Alternatives

Qwen4-Max Reviews

Qwen4-Max

Alibaba
5.0 out of 5 stars
Claude Code Reviews

Claude Code

Anthropic
5.0 out of 5 stars
Muse Spark 1.2 Reviews

Muse Spark 1.2

Meta
5.0 out of 5 stars
Muse Spark Open Reviews

Muse Spark Open

Meta
5.0 out of 5 stars
Hermes Agent Reviews

Hermes Agent

Nous Research
3.5 out of 5 stars
Grok 4.6 Reviews

Grok 4.6

SpaceXSI
5.0 out of 5 stars
Qwen3.8-Max Reviews

Qwen3.8-Max

Alibaba
5.0 out of 5 stars