homeservicesworkaboutblogfree templatescontactFree Tools →Free AI ModelsResearch LibraryROI CalculatorSavings CalculatorAI Readiness ScoreHire vs. AutomateAutomation Quote
book a 30-min call
home / free-models / mistral-large-3

Mistral Large 3 675B

Mistral AI · European Union · Apache 2.0

Commercial use: Yes — free for commercial use

Apache 2.0 on a 675-billion-parameter frontier model from an EU-headquartered lab. Unrestricted commercial use, and the contracting entity for the hosted API is French and directly subject to GDPR.

What this means for your business

The strongest model from a European lab at a size you can genuinely run yourself, under Apache 2.0.

Why it should matter to you

If EU data residency is a hard requirement, this is the shortest path to a defensible answer. Apache 2.0 with no conditions, and if you use the hosted API instead, the entity you contract with is French and directly subject to GDPR — no transfer framework to justify to a DPA.

How it connects to our work

Worth separating two things that get conflated in procurement: EU jurisdiction, EU data residency, and EU-only sub-processors are three different commitments. Ask for all three in writing if all three matter. Self-hosting collapses the question entirely.

From our field notesWhere your server sits is the decision, not whose contract you sign.

Parameters675B
Active per tokendense
Context256K
Modalitytext, code
Memory @ Q4_K_M~380 GB
Memory @ Q8_0~715 GB
LicenceApache 2.0
Last verified2026-09

Advantages

  • Frontier scale under Apache 2.0 from a European lab — the cleanest compliance story available at this capability
  • EU jurisdiction removes the international-transfer question rather than papering over it with a framework
  • NVFP4 and Eagle variants published alongside, so serving is a solved problem

Disadvantages

  • 675B dense is enormous — this is a datacentre model, not a self-host
  • Mistral-Medium-3.5 in the same range ships under a custom "other" licence, so the family is mixed
  • Behind the top Chinese open models on several reasoning benchmarks

Reach for it when

Regulated EU work at frontier capability where both the licence and the jurisdiction have to hold up. Nothing else combines Apache 2.0, EU incorporation and this scale.

Where it falls down

Cost-sensitive or local deployment. If you want European jurisdiction on hardware you own, Mistral Small 4 is the realistic choice.

Running it

Multi-GPU datacentre, or hosted on La Plateforme. See the hardware sizing tables for how that maps to specific chips and cards, and the quantisation guide for what you give up at each bit width.

will it fitMistral Large 3 675B against common GPUs
Mistral Large 3 675B at Q4_K_M
0 GB
Mistral Large 3 675B at Q8_0
0 GB
RTX 3060 12GB
0 GB
RTX 4060 Ti 16GB
0 GB
RTX 4090
0 GB
RTX 5090
0 GB
RTX 6000 Ada
0 GB
A100 80GB
0 GB
H100 80GB
0 GB
Weight sizes for Mistral Large 3 675B as recorded in this directory; GPU memory from the hardware table on the directory hub. Weights only: context adds KV cache.
where the weights fitGPU and Mac memory, checked
Q4_K_MQ8_0
RTX 3060 12GB (12 GB)
RTX 4060 Ti 16GB (16 GB)
RTX 4090 (24 GB)
RTX 5090 (32 GB)
RTX 6000 Ada (48 GB)
A100 80GB (80 GB)
H100 80GB (80 GB)
Mac, 16 GB unified
Mac, 24 GB unified
Mac, 32 GB unified
Mac, 36 GB unified
Mac, 64 GB unified
Mac, 96 GB unified
Mac, 128 GB unified
Mac, 192 GB unified
Mac, 512 GB unified
✓ fits with room for context, ~ fits with under 15% headroom, ✕ does not fit. Weights only, computed from the figures on this page.

Free tiers carrying this model

ProviderTypeThe catchLive limits
Mistral La PlateformeFirst-party APINot every Mistral model is commercially free even when the weights are public — Codestral ships under a non-production licence. Check per model, not per vendor.check →
OpenRouterAggregatorFree-pool membership changes without notice — a model you built on can stop being free, and the list above will drift. Data passes through OpenRouter and then the upstream provider, so check the privacy and routing settings before sending anything sensitive.check →
NVIDIA NIMInference hostIt is a credit grant, not a permanent free tier. When the credits run out you are on a paid plan or self-hosting the container.check →

Jurisdiction: European Union

Mistral is the only EU-headquartered lab shipping frontier-adjacent open weights. Contracting with an EU entity removes the international-transfer question entirely rather than solving it with a framework. Most of the Mistral open range is Apache 2.0, so you can also self-host and remove the vendor from the data path altogether.

Watch for: EU jurisdiction is not the same as EU data residency, and neither is the same as EU-only sub-processors. Ask for all three in writing if it matters to you. Also note that Codestral ships under a non-production licence — being a European vendor does not make every model commercially free.

Frequently asked

Can I use Mistral Large 3 675B commercially?

Apache 2.0 on a 675-billion-parameter frontier model from an EU-headquartered lab. Unrestricted commercial use, and the contracting entity for the hosted API is French and directly subject to GDPR.

What hardware do I need to run Mistral Large 3 675B?

Multi-GPU datacentre, or hosted on La Plateforme. Weights alone are roughly 380 GB at Q4_K_M and 715 GB at Q8_0. Add KV cache on top of that, which grows with your context length.

What licence is Mistral Large 3 675B released under?

Apache 2.0. Full commercial use, modification and redistribution. Patent grant included. The most permissive licence in common use for open-weight models.

Where can I use Mistral Large 3 675B for free?

Free tiers carrying it include Mistral La Plateforme, OpenRouter, NVIDIA NIM. Limits differ per provider and change often, so check each provider's own limits page. You can also self-host the weights, which has no rate limit at all.

Similar models

Wiring Mistral Large 3 675B into something real?

We build the evaluation harness, the failover and the cost ceilings around a model like this, so it survives contact with production.

Maps to AI agent development, AI automation development and AI strategy and consulting. Or see it working: our case studies.