Llama Guard 4 12B.
Llama Guard 4 12B is Meta's 12B safety classifier for text and multi-image prompts and responses, returning safe or unsafe plus the hazard categories broken.
1× Nvidia GeForce RTX 3060 · $0.038/hr.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
Variants in the Llama Guard family.
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 is the cheapest rate on offer now. A multi-card setup uses a real instance of that size when one is rented; otherwise it is the single-card rate times the count.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
48 GB |
$0.27/hr
Vast.ai · on-demand
|
19.6 | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
16 GB |
$0.074/hr
Clore.ai · on-demand
|
26.1 | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
12 GB |
$0.038/hr
Nosana · on-demand
|
42.0 | Compare → |
What it costs per month across providers.
Estimate your monthly bill for Llama Guard 4 12B 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 Llama Guard 4 12B, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month. Power is included in a rental's hourly price.
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 Llama Guard 4 12B.
Llama Guard 4 12B is Meta's safety classifier, not a general assistant. It was pruned from Llama 4 Scout and tuned to label prompts and responses as safe or unsafe, naming the categories broken. It reads text and multiple images. Meta tested it mostly with about three images, warns that prompt injection could bypass it, and says hazards needing current facts, such as elections, need more complex systems. Weights are gated under the Llama 4 Community License.