Run Llama 3.1 8B locally
Llama 3.1 8B is 8B parameters (dense). Here's the VRAM it needs at each quantization, the smallest GPU that fits, and how a self-hosted box compares with the cheapest verified API.
VRAM by quantization
| Quantization | VRAM needed | Fits on | Notes |
|---|---|---|---|
| int4 (Q4) | 4.8 GB | RTX 4090 (24 GB) ($0.34/hr) | smallest footprint, minor quality loss |
| int8 (Q8) | 9.6 GB | RTX 4090 (24 GB) ($0.34/hr) | near-lossless |
| fp16 (full) | 19.2 GB | RTX 4090 (24 GB) ($0.34/hr) | reference quality |
VRAM ≈ params × bytes/param × 1.2 (overhead). See the fullmethod and break-even calculator.
Self-host vs API
Cheapest API for Llama 3.1 8B is $0.06 / 1M tokens(blended) via DeepInfra. Running it yourself in int4 fits aRTX 4090 (24 GB) at $0.34/hr — about $248.2/month at 24/7. Those two lines cross at roughly 4.1B tokens/month: below that the API wins on cost, above it the dedicated GPU does (assuming you keep it busy). Tune your own volume in the break-even calculator.
Run it at home
No cloud account needed: Llama 3.1 8B in int4 (4.8 GB) fits a RTX 5090 (32 GB) + host PC (plus a host PC), a $3,500 one-off buy. Amortized over 3 years that's about $0.1332/hr — hardware only, electricity aside. Against the DeepInfra API at $0.06/1M, buying it pays for itself after roughly 58.3B tokens total. Below that the API is cheaper; a machine that mostly sits idle rarely earns back its price.
Get the weights
Quantized builds on the model hubs — links search live, so they track new builds as they appear:
- GGUF builds llama.cpp · Ollama · LM Studio
- AWQ builds GPU serving (vLLM)
- GPTQ builds GPU serving
- MLX builds Apple silicon
- Ollama library one-command local run
License: open weights under Llama Community License · terms. Compare all providers on the llama-3.1-8b API page.