AMD Releases Instella: Open 3 Billion Parameter Language Models
By
Zac Zuo
More crust than filling. Mostly air.
Summary
AMD has released Instella, a high-performance 3 billion parameter language model trained on MI300X hardware. The model weights are available under a ResearchRAIL license while the code uses an MIT license.
Key quotes
· 3 pulledInstella, from AMD, is the high-performance 3B language models.
ResearchRAIL license for model weights, MIT license for code.
Trained on MI300X.
You might also wanna read
RTP-LLM: Alibaba's High-Performance Inference Engine for Large Language Model Deployment
This paper presents RTP-LLM, a high-performance inference engine developed by Alibaba for industrial-scale deployment of Large Language Mode
arxiv.org·2d ago
Building a Distributed LLM Inference Cluster with AMD Ryzen AI Max+ Systems
This article provides a technical guide on building a distributed inference cluster using AMD's Ryzen AI Max+ AI PC platform to run a one tr
amd.com·3mo ago
Microsoft Releases bitnet.cpp: Official Inference Framework for 1-bit Large Language Models
Microsoft has released bitnet.cpp, an official inference framework for 1-bit large language models (LLMs) like BitNet b1.58. The framework p
github.com·2mo ago
Zyphra's ZAYA1-8B Matches Frontier AI Models on Benchmarks Using Under 1 Billion Active Parameters, Trained on AMD Hardware
Zyphra released ZAYA1-8B, a model that matches or competes with frontier AI models like DeepSeek-R1, Claude Sonnet 4.5, and Gemini 2.5 Pro o
firethering.com·24d ago
Mistral AI Releases Mistral 3: Next Generation of Open-Source Multimodal AI Models
Mistral AI announces Mistral 3, a new generation of open-source multimodal AI models including three small dense models (14B, 8B, 3B) and Mi
mistral.ai·6mo ago
IBM's Granite 4.1: 8B Parameter Open-Source Model Competes with Models Four Times Its Size
IBM released Granite 4.1, a family of open-source enterprise language models (Apache 2.0 licensed) trained on 15 trillion tokens. The stando
firethering.com·1mo ago