Mistral vs Qwen: two open model families, compared

Disclosure first: I build Outlier, a Mac app that runs models on-device, and five of its six tiers are Qwen weights. The sixth is Gemma. I'm not a neutral referee. What I offer instead is the view of someone who shipped one of these families commercially, read the licenses because he had to, and measured what each size costs in RAM.

What each family ships

Mistral AI is French. Its Mistral 3 release in December 2025 put out Mistral Large 3 (675B total, 41B active, sparse MoE) plus Ministral 3 at 3B, 8B and 14B, all Apache 2.0, in base, instruct and reasoning variants. Mistral Small 4 followed: also Apache 2.0, 119B total with 6.5B active across 128 experts, 256k context. Mistral's API price list now groups models as flagship, specialized, third-party hosted and code models and no longer says which are open, so check the license on each model's Hugging Face card. Small 4, Large 3, Ministral 3 at all three sizes and Magistral Small are Apache 2.0. Medium 3.5 and the 123B Devstral 2 are open weights under a Modified MIT License, which Medium's card describes as carrying exceptions for companies with large revenue.

Qwen's open-weight models up to 397B-A17B, and Qwen3.8-27B, list Apache 2.0 on their Hugging Face model cards. The current run: Qwen3.5 at 397B-A17B in February, then 122B-A10B, 35B-A3B and 27B, then 9B, 4B, 2B and 0.8B in March; Qwen3.6 at 35B-A3B and 27B in April; Qwen3.8 in August, with a 2.4T-A95B flagship and a 27B. Qwen3.6-27B is dense, with a native 262,144-token context extending toward a million with RoPE scaling.

The asymmetry that matters isn't the model count. It's the ladder. Qwen gives you one prompt format from 0.8B to a 397B MoE. Mistral gives you a very good small model, a 14B, then a large jump.

Licensing, the part that decides products

Apache 2.0 is a standard OSI license: use, modify, quantize, fine-tune, redistribute, sell. No approval step, no per-seat fee, no acceptable-use appendix bolted on. Mistral's Small 4, Large 3 and Ministral 3 models and Qwen's releases up to 397B-A17B, plus Qwen3.8-27B, sit there: checked on the model cards, not the marketing pages. Not everything does. Mistral's Medium 3.5 and Devstral 2 are under a Modified MIT License, and Qwen3.8-2.4T-A95B ships under its own Qwen3.8-Max License, which asks for the model name to be displayed above 100 million monthly active users or US$20 million in monthly revenue and requires a separate license for large model-as-a-service businesses.

Compare that to the vendor-specific licenses used elsewhere in open weights: Google's Gemma Terms of Use, Meta's Llama Community License. Those aren't OSI licenses. They're vendor documents with their own use policies you have to read, comply with, and pass through to your users. I know because I check it: Outlier's Quick tier is a Gemma 4 model, which Google publishes under Apache 2.0 (see the model card), so all six tiers ship Apache 2.0; I had listed Quick under the Gemma Terms of Use until I read the card. That's why Apache 2.0 keeps winning inside product teams.

What Apache 2.0 does not give you: training data or a recipe. Neither vendor publishes the corpus. "Open weights" means redistribution rights over a binary, not reproducibility.

Mistral and Qwen, size for size

What runs on a Mac, at what RAM

This is where the families diverge most. A 4-bit MLX build of Ministral 3 8B is 5.6 GB, leaving real headroom for KV cache on a 16 GB MacBook Air, and it includes a vision encoder, so it's multimodal at that size.

Above that, Mistral gets awkward on a laptop. "Mistral Small 4" is a name, not a size: 119B total won't fit 16 GB at any usable quantization, even with 6.5B active per token. On a 64 GB Mac Studio that sparse shape is genuinely flattering: activation cost stays small while quality tracks something much larger.

Qwen fills the middle. Qwen3.6-27B in 4-bit MLX is 16.1 GB on disk; in Outlier it's the Core tier, 15.13 GB with a 24 GB floor, 20.7 tok/s on an M1 Ultra. Nano is Qwen3.5-4B, 2.37 GB, 6 GB floor, 71.7 tok/s on the Ultra and about 32 on an M4 Air. Fuller mapping: how much RAM you need for local AI.

Where Mistral genuinely wins

I ship Qwen. Mistral is still the better choice in places I care about.

• The 8B slot. Apache 2.0, multimodal, 256k context, 5.6 GB at 4-bit. For a 16 GB machine that needs vision, that's a hard package to argue with.
• Sparse efficiency at the top. 119B total with 6.5B active is the architecture that suits Apple's unified memory on a high-RAM Mac.
• Jurisdiction and a readable paper trail. Mistral is EU-based, and its privacy policy has a table of what it does with your data, including training on "Your Input and Output, subject to your opt-out." Read the table before you assume anything about a given plan.
• Legible pricing. Flat published per-million rates, a 50% batch discount, a 90% cached-input discount. Model Studio uses volume-tiered pricing on several models and varies by region, which makes an honest quote harder.

Where Qwen genuinely wins

• Apple Silicon as a first-class target. Qwen's README documents mlx-lm and mlx-vlm at release; Mistral's cards lead with vLLM and datacenter GPUs. For a Mac app that's the difference between integrating a release and porting one.
• One ladder, one prompt format. 0.8B, 2B, 4B, 9B, 27B, 35B-A3B, 122B-A10B, 397B-A17B: same family, same chat template, same tokenizer. You can move a product from a MacBook Air to a Mac Studio to a datacenter without rewriting the prompt layer, and the tier you outgrow always has a next rung. Mistral's general-purpose open line goes 3B, 8B, 14B, then straight to 119B.
• The 24–48 GB middle. That gap is the practical one. On a 24 GB Mac, Qwen3.6-27B fits in 4-bit at 16.1 GB and 35B-A3B at about 19 GB. Mistral's general-purpose open catalogue has nothing between the 14B and an 119B sparse model that wants a 64 GB machine. Its other open releases are specialists (Devstral 2 for code, Voxtral for speech) not general chat models.
• Context on a dense local model. 262k native on a 27B you can run at home changes whole-repo work.

Two things I expected to put on that list didn't survive checking, so they aren't on it. Ollama carries Ministral 3 as an official library entry (3B, 8B and 14B, 13 tags, 1.5M downloads, vision and tool use) so distribution is not a Qwen-only advantage; the only hole is Mistral Small 4. And Mistral's Vibe CLI is Apache 2.0, documents an offline local-models path, and lists a Mistral account as optional, so "Qwen's tooling is the local-friendly one" is not true as a general claim. I'd rather run a list where every item checks out than pad the column.

Which family to build on

Pick Mistral if you're targeting 16 GB machines and need vision in a small model, if EU jurisdiction is a procurement requirement, or if you want flat published API pricing.

Pick Qwen if you're shipping on Apple Silicon, if you need one family to cover a MacBook Air and a Mac Studio without changing prompt formats twice, or if long context on a dense local model decides your product.

Don't assume the frontier stays closed. github.com/QwenLM/Qwen3.7 returns a 404, but the Qwen repository's README now covers Qwen3.8, including an open 2.4T-A95B release it calls Qwen-Max-class; on Mistral's side, Medium 3.5 is open weights under a Modified MIT License. Both move fast enough (Qwen shipped 3.5, 3.6, an API-only 3.7 and an open 3.8 in six months) that bundling weights means a permanent re-quantize treadmill. That's my week.

One last note. I could not verify the consumer Qwen chat app's data handling, those pages render client-side and returned no policy text. Don't accept a claim about Qwen Chat without a primary read. If you'd rather skip the question, that's the case for running weights yourself.