AI models · Mistral AI

Mistral Large 3

Mistral Large 3 came out in December 2025, and until the Large 4 weights are published it is the largest Mistral model you can download. It has 675 billion parameters, 41 billion of them active per answer, and a permissive Apache 2.0 license. It is much cheaper in the API than the newer Large 4, and Mistral has not set an end date for it yet.

Strengths

  • The largest Mistral model with downloadable weights under the permissive Apache 2.0 license, until the Large 4 weights ship.
  • Cheap for a large model: $0.50 per million input tokens and $1.50 per million output tokens.
  • A 256K token context window, and it understands both text and images.

Weaknesses

  • It is being replaced by Mistral Large 4, in public preview since October 2026.
  • Running 675 billion parameters yourself takes a server with several powerful GPUs.

Best for

  • self-hosting a large open model with a permissive license
  • low-cost API use with data kept in the EU

Not ideal for

  • teams that can wait for the final Large 4

How to use it in Europe

You can use it through the Mistral API across the EU, and data is processed in the EU by default. The weights are available under Apache 2.0, but you need a powerful server to run them. It suits anyone who needs a stable large model now rather than waiting for the final Large 4.

Price compared with similar models

1 3 10 30 Beam 1.01 Mistral Large 31.5 DeepSeek V4 Pro3.96 Mistral Large 44.18 Qwen3.8-Max6 Qwen3.8-2.4T-A95B6 Grok 4.76 Gemini 4 Carbon10 Gemini 4 Argon10 Kimi K3.1 Pro10 Kimi K315 Claude Opus 5.520 Muse Wonder20 GPT-6 Astra50
Output price in USD per 1M tokens (log scale). Lighter bars are open-weight models.

More models from this vendor

  • Mistral Large 4: Mistral's largest model, still in preview. Built for demanding tasks, long documents and security analysis with data kept in the EU.
  • Mistral Medium 3.5: Mistral's mid-size open-weight model and the default in the Vibe app. Good for everyday work and coding with data kept in the EU.
  • Mistral Small 4: Mistral's cheap, fast open model under Apache 2.0 for everyday tasks through the API or on your own server.
  • Ministral 3 14B: Mistral's small open model under Apache 2.0 that runs on a single GPU, even without an internet connection.
  • Codestral 25.08: Mistral's model for code completion right in your editor. Fast and cheap, but dated for autonomous coding agents.

Similar models from other vendors

  • Claude Opus 5.5 (Anthropic): Anthropic's main model for demanding work, coding and long documents, and the one Anthropic recommends starting with.
  • DeepSeek V4 Pro (DeepSeek): DeepSeek's largest open-weight model for demanding text tasks at a low price. It does not understand images.
  • Gemini 4 Carbon (Google): Google's most capable model, announced on 28 October 2026. So far only selected cybersecurity experts have access.
  • Gemini 4 Argon (Google): Google's most capable publicly available model, out in October 2026. Built for demanding analytical work where accuracy matters and made-up answers are costly.
  • Muse Wonder (Meta): Meta's strongest model for long-running agents and coding, still released as a preview.
  • Kimi K3.1 Pro (Moonshot AI): Moonshot AI's newest model from October 2026, built for agentic coding, office documents and very long outputs.

Sources

Prices and models change often. The data comes from official vendor pricing pages and documentation, and every page shows when it was last checked. Before a larger purchase, confirm the price with the vendor.

← All AI models

Not sure which model fits your business?

I can help you pick a model and plan based on what you actually need, set it up safely for company data and train your team to use it well. Let’s start with an intro consultation.

See pricing jozef@walterstein.eu