Local LLMs
TurboFieldfare: Run Gemma 4 26B in 2GB RAM on Apple Silicon
July 29, 2026 2 min read
AI-assisted · editor-reviewedHow we use AI
TurboFieldfare is a custom Swift and Metal runtime that runs the 26-billion-parameter Gemma 4 model on Macs with limited RAM. By streaming experts from SSD instead of loading the entire model, it achieves efficient inference on 8 GB Macs.
Impact: High
Why it matters
It allows running massive models on low-memory hardware.
TL;DR
- 01Runs 26B model in ~2GB RAM budget.
- 02Uses Swift and Metal.
Key facts
- RAM Requirement
- ~2 GB
Architecture and SSD-Streaming
At each transformer layer, Metal computes attention and the router from resident weights. The CPU uses the router's top-8 expert IDs to plan against the layer's 16-slot LFU cache, then fills misses with bounded parallel pread calls into Metal-visible buffers. Metal computes the resident shared-expert branch while those reads run, then combines the shared and routed outputs.
#TurboFieldfare#Swift#Metal