Llama Guard 4 12B
meta-llama/llama-guard-4-12b
Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.
blended = 90.9% input × [(1 − 70% cache) × input + 70% cache × cache-read] + 9.1% × output + cache writes (30% of uncached input) where a write price is published
A provider that publishes no cache price is blended with cache share 0 (all input at the input price) and labelled “no cache price published” — never penalised against identical in/out prices.
No published pricing for this dimension.
| Provider | Input | Output | Cache rd | Blended |
|---|---|---|---|---|
Kilo Gateway router / aggregator | $0.180 | $0.180 | — | $0.180no cache price published |
OpenRouter router / aggregator | $0.180 | $0.180 | — | $0.180no cache price published |
Free tiers (rate-limited) · 1
No payment required; rate limits and quotas apply (not tracked yet). Excluded from cheapest, median and spread.
- Nvidia meta/llama-guard-4-12bNVIDIA developer account + API keyfree tier
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.