Mistral Small (latest)
MistralOpen weights
Mistral Small (latest) by Mistral costs $0.15 per 1M input tokens and $0.60 per 1M output tokens, with a 262K-token context window.
Pricing
Input (per 1M tokens)
$0.15
Output (per 1M tokens)
$0.60
Cached input (per 1M)
$0.02
Specifications
- Provider
- Mistral
- Context window
- 262K tokens
- Parameters
- —
- Released
- Mar 2026
- Open weights
- Yes
- Frontier model
- No
Benchmarks
- GPQA Diamond
- 47.5
- LMArena Vision Elo
- 1141.6
Speed & latency
Time to first token
0.54s
Output speed
254 tok/s
Fastest host: mistral (via vercel-ai-gateway).
Compare Mistral Small (latest) with…
Mistral Small (latest) vs Mistral: Voxtral Small 24B 2507$0.30/1MMistral Small (latest) vs Mistral: Mistral Large 4$2.09/1MMistral Small (latest) vs Mistral: Mistral Medium 3.5$7.50/1MMistral Small (latest) vs Mistral: Mistral Medium 3.5 (batch)$3.75/1MMistral Small (latest) vs Mistral: Voxtral Mini TTS$0.00/1MMistral Small (latest) vs Mistral: Mistral Large 3 2512$1.50/1M
FAQ
How much does Mistral Small (latest) cost?
Mistral Small (latest) is priced at $0.15 per 1M input tokens and $0.60 per 1M output tokens. Output tokens cost more than input.
What is Mistral Small (latest)'s context window?
Mistral Small (latest) supports a context window of 262K tokens — the maximum prompt plus response length it can handle in a single request.
Is Mistral Small (latest) open weight?
Yes — Mistral Small (latest) has openly available weights.
Data as of 2026-10-14. Source: models.dev (open). Pricing is per 1M tokens (USD); confirm with the provider before production use.