AI BriefingOllama Adds Direct llama.cpp Support
·2026.05.14 23:33
Key point
Ollama has switched from a GGML-based build to direct llama.cpp support.
Details
Ollama v0.30.0-rc15 has switched from a GGML-based build to direct llama.cpp support.
This change enables GGUF file format compatibility, and on Apple Silicon, model inference is accelerated with MLX.
Pre-release feedback items are as follows.
- Performance improvements or degradations
- Errors or crashes that were not present before
- Changes in memory usage
Known unsupported models are as follows.
laguna-xs.2llama3.2-vision
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Underdog Releases 'Husky', a Model-Specific Inference Engine Up to 4.5x Faster than MLX
Jared Palmer Releases 'Kev', a Lightweight Decision-Making Model Based on Qwen3.5