The creators of the model actually have their own platform where you can access this model and others via API: https://console.mistral.ai/

I just discovered Groq, which does 485.08 T/s on mixtral 8x7B-32k

No idea on pricing but supposedly one can email to api@groq.com

OpenRouter is generally a good option (already mentioned), the best part is that you have a unified API for all LLMs, and the pricing is the same as with the providers themselves. Although for OpenAI/Anthropic models they were forced (by the respective companies) to enable filtering for inputs/outputs.

Both already mentioned, but I am using Anyscale Endpoints with great success, very fast and will work on ten jobs at a go out of the box. Together.ai also seems to work fine in my initial tests, but haven't tried it at scale yet.

I have used both Mistral’s commercial APIs and also AnyScale’s commercial APIs for mixtral-8-7b- both providers are easy to use.

I also run a 3 bit quantization of mixtral-8-7b on my M2 Pro 32G memory system and it is fairly quick.

It is great having multiple options.

openrouter, fireworks, together.

we use openrouter but have had some inconsistency with speed. i hear fireworks is faster, swapping it out soon.

I work for Groq and we serve the fastest available version of Mixtral (by far) and we also have a web chat app. I'll refrain from linking it because it has already been linked and I don't want to spam, but I'm available to answer any questions people have about Groq's hardware and service.

Together.ai seems to be the best, incredibly fast.

I personally like Anyscale Endpoints

I've had good experiences with Together, and they have very competitive pricing.

You also have Replicate: https://replicate.com/mistralai/mixtral-8x7b-instruct-v0.1