ram-coffers
LLM infrastructure cost reduction via NUMA-aware weight banking: 147 t/s (8.8x stock llama.cpp) on refurbished enterprise POWER8. Self-hosted inference, no cloud APIs. Part of the Proof of Physical AI stack.
Scottcjn/ram-coffers
View on GitHub ↗LLM infrastructure cost reduction via NUMA-aware weight banking: 147 t/s (8.8x stock llama.cpp) on refurbished enterprise POWER8. Self-hosted inference, no cloud APIs. Part of the Proof of Physical AI stack.
ram-coffers grew +5.5% in the last 30 days, ranks 198th of 303 in Inference Optimization, actively maintained, last commit 11 days ago.
Star trajectory
81 days tracked+5.5% in 30d, above the Inference Optimization median of +1.5%.
Last commit 11 days ago, fresher than the median of 15d.
How it compares in Inference Optimization
Ranked 198th of 303 by stars.
- Language
- Python
- License
- GNU Affero General Public License v3.0
- Contributors
- 36
- Forks
- 61
- Watchers
- 30
- Open issues
- 14
- Releases
- 0
- Created
- 2026-01-19
- Stars this week
- +2
You've seen where ram-coffers stands today.
Get the moves before everyone else: the 5 Inference Optimization repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.