Mistral Large 4 Pricing

Mistral Large 4 API pricing

Mistral Large 4 API pricing is $1.36 per million input tokens and $4.18 per million output tokens, verified October 8, 2026.

Input (per 1M tokens)

$1.36

Output (per 1M tokens)

$4.18

Cached input (per 1M tokens)

$0.14

Context window

1,000,000 tokens

No Mistral Large 4 price changes have been recorded yet.

No price snapshots recorded yet.

Frequently asked questions

Short answers to the most common questions about Mistral Large 4 pricing, limits, and capabilities.

Notes

Added 2026-10-08. Announced 2026-10-06 (https://mistral.ai/news/mistral-large-4/), public preview, open-weight. 1.05T total parameters / 52B active, granular MoE, natively multimodal with a 1.6B vision encoder, 1M context. Weights to be released end of October 2026. STORED RATES ARE LIST PRICE: $1.36 input / $0.14 cached input / $4.18 output per M tokens, read from https://docs.mistral.ai/inference/pricing on 2026-10-08. PROMO IN EFFECT: Mistral is showing a 50% launch discount during public preview — $0.68 input / $0.07 cached / $2.09 output. Per SPH policy we store the standard list price, not the promotional rate. Re-check when the preview ends; if the sale price becomes the standard price, that is a real price movement and needs a snapshot row. SOURCE NOTE: https://docs.mistral.ai/inference/pricing is a clean per-token table for every Mistral model and should replace https://mistral.ai/pricing/ as the provider URL in the AI Models refresh allowlist — the marketing page has repeatedly failed to expose versioned rates. The docs table also confirms Mistral Large 3 at $0.50/$1.50 (still Live, not superseded), Medium 3.5 at $1.50/$7.50 and Small 4 at $0.15/$0.60, all matching stored values.