Sao10K: Llama 3 8B Lunaris.
Lunaris is a Llama 3 8B merge by Sao10K for general use and roleplay, built on Llama 3 8B Instruct plus four other models, under the Llama 3 licence.
1× Nvidia GeForce GTX 1660 Super · $0.028/hr.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
24 GB | $0.15/hr | 40.9 | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
16 GB | $0.019/hr | 28.0 | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
6 GB | $0.028/hr | 58.8 | Compare → |
What it costs per month across providers.
Estimate your monthly bill for Sao10K: Llama 3 8B Lunaris across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
Rent the GPU instead of paying per token.
For an open-weights model like Sao10K: Llama 3 8B Lunaris, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
About Sao10K: Llama 3 8B Lunaris.
Lunaris is a Llama 3 8B merge by Sao10K, meant as a generalist and roleplaying model. It is built on Meta's Llama 3 8B Instruct and blends four other models, including Stheno v3.2, using the TIES merge method. The author chose the parts from personal experience and says they balance creativity with better logic. It uses the Llama 3 Instruct template and is released under the Llama 3 licence.