Instella MoE
AMD's 16B parameter MoE model optimized for agentic reasoning on ROCm hardware
About
Instella MoE is AMD's open-source mixture-of-experts language model featuring 16 billion parameters with 3 billion active experts per token. Specifically optimized for AMD's ROCm platform, it delivers enhanced reasoning capabilities for agentic workflows while maintaining computational efficiency through its sparse expert activation. Designed as a foundation for building autonomous agents that require strong logical reasoning and decision-making abilities.
Details
| Type | coding-agent |
| Deployment | self-hosted |
| Supported Models |
Tags
Quick Info
- Organization
- AMD
- Pricing
- open-source
- Free Tier
- Yes
- Updated
- Aug 5, 2026
Also in Agents
AI Data Analysis Agent
Autonomous agent that analyzes datasets and generates visual insights
AI Deep Research Agent
Autonomous agent that conducts comprehensive multi-source research investigations
AI Journalist Agent
Autonomous agent that researches topics and writes structured news articles