Mistral Medium 3.5
mistralai/mistral-medium-3.5-128b
spread (output, p90 ÷ p10)
1.0×
$6.90 – $10.26 · 4 providers · max÷min 1.5×
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 |
|---|---|---|---|---|
Merge Gateway router / aggregator | $0.40 | $2.00 | — | $0.55no cache price published |
Pioneer unclassified | $0.40 | $2.00 | $0.40 | $0.58 |
Requesty router / aggregator | $0.44 | $2.20 | $0.44 | $0.60 |
evroc inference host | $1.73 | $6.90 | — | $2.20no cache price published |
Vercel AI Gateway router / aggregator | $1.50 | $7.50 | $0.15 | $1.19 |
Eden AI router / aggregator | $1.50 | $7.50 | $0.15 | $1.19 |
Mistralfirst-party first-party lab | $1.50 | $7.50 | — | $2.05no cache price published |
NanoGPT router / aggregator | $1.50 | $7.50 | $0.75 | $1.57 |
OpenRouter router / aggregator | $1.50 | $7.50 | — | $2.05no cache price published |
Scaleway inference host | $1.50 | $7.50 | — | $2.05no cache price published |
Cortecs router / aggregator | $1.53 | $7.84 | $0.15 | $1.23 |
GreenPT inference host | $2.05 | $10.26 | — | $2.80no cache price published |
Free tiers (rate-limited) · 2
No payment required; rate limits and quotas apply (not tracked yet). Excluded from cheapest, median and spread.
- Nvidia mistralai/mistral-medium-3.5-128bNVIDIA developer account + API keyfree tier
- Kenari mistral-medium-3-5:freefree tier
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.