Mistral pricing · auto-updated daily

Mistral API pricing — June 12, 2026

All 12 Mistral text models with token prices ($ per 1M), context window and max output, refreshed daily from the official pricing source. Bring your Mistral key to flo2 and you pay these exact prices — zero markup — with fallback, racing and per-request cost accounting on top.

Cheapest output
$0.1/M
ministral-3b-latest
Cheapest input
$0.1/M
voxtral-small-latest
Biggest context
tok
mistral-medium-latest
Models
12
tracked daily
ModelContextMax outIn $/MCached inOut $/MReasoning
mistral-medium-latest $1.5 $7.5
magistral-medium-latest $2 $5
devstral-medium-latest $0.4 $2
magistral-small-latest $0.5 $1.5
mistral-large-latest $0.5 $1.5
codestral-latest $0.3 $0.9
voxtral-small-latest $0.1 $0.4
devstral-small-latest $0.1 $0.3
mistral-small-latest $0.1 $0.3
ministral-14b-latest $0.2 $0.2
ministral-8b-latest $0.15 $0.15
ministral-3b-latest $0.1 $0.1

Prices in USD per 1,000,000 tokens, fetched 2026-06-12 from the official Mistral pricing source. Verify before large commitments. Click any column header to sort.

More providers: OpenAI · Anthropic · Google Gemini · xAI Grok · Groq · Cerebras · DeepInfra · OpenRouter · NVIDIA NIM · or the full cross-provider comparison.

Use Mistral through one key — zero markup.
flo2 routes every call to the cheapest, fastest model that clears your bar, with fallback, racing and true cost accounting. Free during Beta.
Get your flo2 key →
© 2026 flo2.com — the zero-markup LLM gateway & router. blog · all providers · flow → to