Magistral Small
mistralai/magistral-small-2506
spread (output, p90 ÷ p10)
— too few providers
$1.50 – $1.50 · 2 providers · max÷min 1.0×
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.
| Provider | Input | Output | Cache rd | Blended |
|---|---|---|---|---|
IO.NET inference host | $0.50 | $1.50 | $0.25 | $0.51 |
Mistralfirst-party first-party lab | $0.50 | $1.50 | — | $0.59no cache price published |
Pioneer unclassified | $0.50 | $1.50 | $0.50 | $0.63 |
Free tiers (rate-limited) · 1
No payment required; rate limits and quotas apply (not tracked yet). Excluded from cheapest, median and spread.
- Nvidia mistralai/magistral-small-2506NVIDIA developer account + API keyfree tier
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.