Skip to content
Arm
Git
Search or jump to…
Ctrl K
Categories
Theme: System
Sign in
Get started
armgit
/
turbo-fieldfare
Public
Imported from GitHub
Fork
0
Watch
0
Star
0
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
Swift
apple-silicon
gemma
gemma4
gemma4-26b-a4b
gpgpu
llm
llm-inference
local-ai
macos
metal
on-device-ai
on-device-llm
swift
Code
Commits
Issues
Pull requests
Wiki
Builds
Discussions
Projects
Insights
Search
Files
main
3 branches · 24 tags
Go to file
t
Code
turbo-fieldfare
/
docs
/
experiments
/
summaries
Name
Last commit message
Last commit date
..
MD
01-model-install-and-expert-io.md
8.4 KB
MD
02-decode-moe-int4-and-router.md
12.5 KB
MD
03-expert-cache-prediction-and-layout.md
6.7 KB
MD
04-rdadvise.md
5.6 KB
MD
05-attention-and-kv-cache.md
12.0 KB
MD
06-prefill.md
12.7 KB
MD
07-fusions-head-and-orchestration.md
10.3 KB
MD
08-sampling-tokenization-and-output.md
4.0 KB
MD
09-validation-and-measurement-lessons.md
8.9 KB
MD
10-long-context.md
1.6 KB