AI Rate CardLlama-Guard-3-8B vs Llama Guard 4 12B: API price and specs
Llama-Guard-3-8B is 62% cheaper than Llama Guard 4 12B on input price.
Llama-Guard-3-8B is 79% cheaper than Llama Guard 4 12B on output price.
Llama-Guard-3-8B is 65% cheaper than Llama Guard 4 12B on a 1M-in / 200k-out workload.
Llama Guard 4 12B has a 25% larger context than Llama-Guard-3-8B.
| Field | Llama-Guard-3-8B | Llama Guard 4 12B |
|---|---|---|
| Lowest input, per 1M | $0.055 | $0.14 +161.8% |
| Median input, per 1M | $0.27 +49.7% | $0.18 |
| Lowest output, per 1M | $0.03 | $0.14 +380% |
| Median output, per 1M | $0.0425 | $0.18 +323.5% |
| Lowest unit price | — | — |
| Cost of 1M in + 200k out | $0.061 | $0.17 +183.3% |
| Context | 131K +20% | 164K |
| Max output | 131K | 131K |
| Providers | 3 +62.5% | 8 |
| Regions | GLOBAL, US | US |
| Accepts | text | image, text |
| Returns | text | text |
| Type | Moderation | Moderation |
| Free option | no | yes |
| Cheapest provider | deepinfra | kilo |
Your own workload: how many tokens the model gets (input) and gives back (output) - for example, in a month. The "Cost" row of the table above is recalculated for these numbers, for each model.
Computed from current offers each time the page opens; the best value of a row is bold, the percent is how far behind the best.
Providers selling both
Frequently asked questions
Which is cheaper, Llama-Guard-3-8B or Llama Guard 4 12B?
Llama-Guard-3-8B is 62% cheaper than Llama Guard 4 12B on input price. Llama-Guard-3-8B is 79% cheaper than Llama Guard 4 12B on output price. Llama-Guard-3-8B is 65% cheaper than Llama Guard 4 12B on a 1M-in / 200k-out workload.
Which has the larger context, Llama-Guard-3-8B or Llama Guard 4 12B?
Llama-Guard-3-8B: 131K tokens; Llama Guard 4 12B: 164K tokens.
Where can I use both Llama-Guard-3-8B and Llama Guard 4 12B?
deepinfra, edenai