Skip to content
RankShark

Mixtral 8x22B Instruct

Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b).

ModelOpen sourceby Mistralsince 2024
26.9

SharkScore

#358 in Models

Why it scores 26.9

Mixtral 8x22B Instruct has a SharkScore of 26.9/100, ranked #358 of 608 in Models (updated 2026-11-19).

Each part compares Mixtral 8x22B Instruct with other listings in Models. Evidence covers 87% of the score's weight.

Adoption
10 · 30%
User reviews
no data · 25%
Maintenance
1 · 20%
Trust & transparency
80 · 15%
Momentum
no data · 10%

Paying never changes this score. How ranking works · method v2026-11-18.2

RankShark review of Mixtral 8x22B Instruct

Automated evaluation by RankShark · 18 Nov 2026 · method v1.0 · How we review

We were not paid for this review.

As of 2026-11-18, Mixtral 8x22B Instruct is an instruction-tuned text model with a 65,536-token context window and OpenRouter pricing of $2 per million input tokens and $6 per million output tokens. [E1][E2][MEASURED]

Mixtral 8x22B Instruct is Mistral's instruction-fine-tuned version of Mixtral 8x22B. Its model card includes inference examples and tool-calling instructions; OpenRouter lists it for API access. [E1][E2]

Best for

  • Developers who want to run the model with Hugging Face Transformers, vLLM, or Mistral's inference tooling. [E1] · model card
  • Applications that need model tool-calling support. [E1] · model card
  • API users who can work within the listed context and token prices. [E2][MEASURED] · OpenRouter

Not for

  • Applications that require image output; OpenRouter lists the output modality as text. [E2] · OpenRouter

Key features

  • Instruction fine-tuning. The vendor describes the model as an instruct fine-tuned version of Mixtral-8x22B-v0.1. [E1] · model card
  • Tool calling. The vendor provides tool-calling examples and documents special tokens for tool calls and results. [E1] · model card
  • Context and completion limits. The OpenRouter listing gives a 65,536-token context length and a 52,428-token maximum completion. [E2] · OpenRouter
  • Input and output modalities. The OpenRouter listing specifies text and file input, with text output. [E2] · OpenRouter
  • Inference options. The model card includes examples using Transformers, vLLM, and Mistral inference tooling. [E1] · model card

Pros

  • The model card provides example workflows for multiple inference libraries. [E1] · model card
  • The model card documents function calling and tool-call tokens. [E1] · model card
  • The Hugging Face model card lists the Apache-2.0 license. [E1] · model card

Cons

  • The OpenRouter listing's maximum completion is 52,428 tokens, less than its 65,536-token context length. [E2] · OpenRouter
  • No free plan or free trial stated on the pages we checked.
  • Whether customer data is used for model training is not stated on the pages we checked.

Pricing

PlanPriceLimits
OpenRouter model listing$2 per million input tokens; $6 per million output tokens. [E2][MEASURED]65,536-token context length; 52,428-token maximum completion. [E2]

Checked 18 Nov 2026

Free plan
Not stated (checked 18 Nov 2026)
Free trial
Not stated (checked 18 Nov 2026)
Trains on customer data
Not stated (checked 18 Nov 2026)
Data retention
Not stated (checked 18 Nov 2026)
Data regions
Not stated (checked 18 Nov 2026)
Data processing agreement
Not stated (checked 18 Nov 2026)
AI models used
Mixtral-8x22B-Instruct-v0.1, an instruct fine-tuned version of Mixtral-8x22B-v0.1. [E1] · model card
API
OpenRouter lists chat completions, completions, messages, and responses endpoints. [E2] · OpenRouter
SDKs
The model card includes Python examples using mistral_common, mistral_inference, and Hugging Face Transformers. [E1] · model card
Webhooks
Not stated (checked 18 Nov 2026)
Platforms
The model card shows inference examples for Hugging Face Transformers, vLLM, and Mistral inference tooling. [E1] · model card
Security claims
No security certification stated (checked 18 Nov 2026)

What we measured

Context window (OpenRouter listing) · now
65,536 tokens
Price per million tokens (OpenRouter listing) · now
$2 in, $6 out
Package downloads · 30 days
37,887
How we tested · sources
  • Hourly uptime and response-time checks of the product's URL.
  • Repository, release and download data from GitHub, npm, PyPI and Hugging Face where they exist.
  • We did not sign up, pay, or use the product as a customer.
  • 18 Nov 2026: Plans or prices changed on the pricing page.
  • 18 Nov 2026: Plans or prices changed on the pricing page.

Something wrong? Report an error with a public link that shows it; corrections re-run the review.

Top alternatives to Mixtral 8x22B Instruct

See all
  • 67.1

    Qwen3.8 27B is an open-weight dense vision-language model from Qwen.

    ModelOpen source
  • 65.5

    Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per t

    ModelOpen source
  • 64.7

    Qwen3.8 Flash is a multimodal reasoning model from Alibaba.

    ModelOpen source
  • 64.3

    Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026.

    ModelOpen source