Question 1

What is Mixtral 8x22B Instruct?

Accepted Answer

Mixtral 8x22B Instruct is a model from Mistral, released on Apr 17, 2024. The Mixtral-8x22B-Instruct-v0.1 Large Language Model (LLM) by Mistral AI is a fine-tuned version of Mixtral 8x22B with 39B active parameters. The model is designed to offer enhanced instruction fine-tuning features and can generate responses fluently in English, French, Italian, German, and Spanish.

It excels in math, coding, reasoning tasks, offers a large context length of 64k, and supports multiple languages. This model provides cost-efficient performance and showcases strong abilities across various language tasks.

Its architecture includes unique characteristics like Grouped-Query Attention and Sliding-Window Attention. Its application extends to diverse areas of text generation, coding assistance, and reasoning tasks.

Question 2

How much does Mixtral 8x22B Instruct cost?

Accepted Answer

On the Respan gateway, input starts at $2.00 per 1M tokens on OpenRouter, and output starts at $6.00 per 1M tokens on OpenRouter, after gateway discounts.

Question 3

Which providers serve Mixtral 8x22B Instruct on Respan?

Accepted Answer

Mixtral 8x22B Instruct is served by OpenRouter.

Question 4

What is Mixtral 8x22B Instruct's context window?

Accepted Answer

Mixtral 8x22B Instruct accepts up to 66K tokens of context.

Question 5

How do I use Mixtral 8x22B Instruct with the Respan gateway?

Accepted Answer

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "mixtral-8x22b-instruct". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Question 6

How fast is Mixtral 8x22B Instruct?

Accepted Answer

Over the last 30 days of requests through the Respan gateway, the fastest provider was OpenRouter at 33 tokens per second, and the lowest response latency was 0.4 seconds on OpenRouter. Speed varies with request size and provider.

Mixtral 8x22B Instruct

Providers

Pricing

Cost by volume

Throughput and latency

Output speed

Response latency

Certified where it counts.

Apps