File 002 · Local Intelligence
Record 026 · Models · 2024
24.26.LLlama 3
Meta
“Llama 3 8B/70B became the new default GGUF on every homelab.”
The 8B that replaced the 7B.
Llama 3 (and 3.1/3.2/3.3) reset local baselines: 8B that felt like last year's 70B for a lot of chores, 70B that fit in 48GB quantized.
The second-wave weights the current local stack assumes.
Filed notes
- Llama 3 8B/70B
- 128K context later 3.1
- Edge 3.2 1B/3B
SIC
7372
Prepackaged Software
NAICS
513210
Software Publishers
