Skip to content

Mistral launches Large 4 preview at half its listed API price

The 1-trillion-parameter mixture-of-experts model ships with a 1M token context window, open weights due by the end of October, and preview pricing roughly 50% below its stated full price.

Source: Mistral AI

Mistral released a public preview of Large 4 on 6 October: a mixture-of-experts model with roughly 1 trillion total parameters and 49 billion active, a 1 million token context window, and native multimodal input. It was trained from scratch on 3,800 Nvidia Grace Blackwell GPUs in Mistral’s own European data centres, on a dataset spanning more than 160 languages, including every official EU language.

The preview API is live now in Mistral Studio at a discount: $0.68 per million input tokens and $2.09 per million output, against a listed full price of $1.36 and $4.18. Mistral says the model leads open models on security-relevant evaluations, scoring 93% on Cybench and 82% on the AA Cyber Index’s vulnerability-reproduction test, and also tops its AutomationBench agentic benchmark. Open weights are due by the end of October.

Why it matters: a 1M token context window at half-price preview rates, with open weights landing within weeks, is worth a benchmark run against whatever you currently use for agentic or coding workloads, especially if European data residency or an eventual self-hosted option matters to you.

The caveat: the discounted pricing is explicitly a preview rate, and Mistral has not said when, or whether, it reverts to the full price.

Share

Related reading

ML for Web Devs

Tokens, context windows, and why your bill is what it is

The unit of billing for language models is not the request or the word. Once you understand tokens and how a context window fills up, the invoice stops being a surprise.

ML for Web Devs

RAG is just search with extra steps

Retrieval-augmented generation gets written up like an architecture. It is a search query, a string concatenation and one API call, and the search query is the part that decides whether it works.

ML for Web Devs

A prompt is a spec, so write it like one

Prompt engineering has very little to do with magic words. It is the same skill as writing a clear ticket for a capable contractor who will not ask you any questions.

More from The Wire

All briefs

Anthropic cuts distillation attacks on Claude by up to 99%

New safeguards, including a reasoning classifier and stricter account limits, now block most attempts to copy Claude by training on its output. Blocked requests in high-confidence cases are billed, and some integrations may need updating.

via Anthropic