Language modelOpen source

Mixtral 8x22B

Mistral AI

Released: April 2024
Data date: September 8, 2026

Price: $2 · $6
Input · Output / 1M tokens
Input: Text
Output: Text

What is distinctive about Mixtral 8x22B is its large sparse mixture-of-experts design. Mistral describes 141 billion total parameters, 39 billion active parameters, and a 64,000-token context window; only part of the eight experts activates per token.

The open weights use Apache 2.0. Mixtral 8x22B extends the MoE architecture of the smaller Mixtral 8x7B to 141 billion total parameters.

Specifications and access

Specification	Value and source
Weights	Open weightsSource
Access	API, LocalSource
Parameters	141BSource
Active parameters	39BSource
Context window	65,536 tokensSource
Notes	8 experts x 22B eachSource
License	Apache-2.0Source
Release	2024-04-17Source
Parameters	141 BSource
Active parameters	39 BSource
Architecture	sparse mixture-of-expertsSource
Input	TextSource
Output	TextSource
Languages	English, French, Italian, German, SpanishSource
Tool calling	YesSource
Structured output	YesSource

Published benchmarks

MedQA

Overall

1.o196.52%

2.Gemini 3.1 Pro Preview96.37%

3.GPT-596.32%

4.GPT-5.496.09%

5.GPT-5 mini96.06%

5.o396.06%

7.o4-mini96.02%

8.Gemini 3 Flash Preview95.81%

9.o3-mini94.83%

10.Grok 4.20 Reasoning94.55%

11.Kimi K2.594.37%

12.GLM-594.27%

13.GLM-4.793.74%

14.GPT-5 nano93.26%

15.Claude Opus 4.593.16%

16.Kimi K2 Thinking92.59%

17.MiniMax M2.592.53%

18.GLM-4.692.23%

19.Claude Sonnet 4.692.06%

20.GPT OSS 120B91.36%

21.GPT-4.191.18%

22.MiniMax M2.191.16%

23.DeepSeek-R190.8%

24.GLM-4.589.98%

25.DeepSeek-V3.289.45%

26.GPT-4o88.16%

27.GPT-4.1 mini84.63%

28.Kimi K2 071183.98%

29.GPT OSS 20B82.88%

30.Mistral Large 382.23%

31.DeepSeek-V3 032482%

32.DeepSeek-V380.9%

33.Command A80.55%

34.GPT-4o mini72.44%

35.Mistral Small 3.169.1%

36.GPT-4.1 nano68.23%

37.Mixtral 8x22B62.14%

38.Llama 4 Scout50.9%

39.Command R+2.65%

higher is betterAs of September 29, 2026Scale: 0% - 100%

Source: Vals AI

gradually.ai

Methodology and context

Comparison cohort: vals-medqa:1:overall
Metric: accuracy
Scale: 0-100
Participants: 39
Methodology and context: Source model ID: together/mistralai/Mixtral-8x22B-Instruct-v0.1. Source model name: Mixtral 8x22B. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-04-16. Latency in seconds: 8.946. Cost per task in USD: 0.000918. Source ID: vals-ai. Evidence capture ID: vals-medqa-2026-09-29-2886a92806d7. Source payload hash: 2886a92806d7b2e892c96203b7c7b462a4e28ec54dae900401734c38aa443883. Parser version: vals-astro-v2. Imported at: 2026-09-29

Source: Vals AI

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

GSM8K maj@8 (Show measurement, test conditions, and source)

Source value: 90.8
Score: 90.8
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; maj@8, as stated in release post

Source Mistral AI release post

Math maj@4 (Show measurement, test conditions, and source)

Source value: 44.6
Score: 44.6
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; maj@4, as stated in release post

Source Mistral AI release post

MMLU (Show measurement, test conditions, and source)

Source value: 77.75
Score: 77.75
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

HellaSwag (Show measurement, test conditions, and source)

Source value: 88.5
Score: 88.5
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

WinoGrande (Show measurement, test conditions, and source)

Source value: 84.7
Score: 84.7
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

ARC-C5 (Show measurement, test conditions, and source)

Source value: 91.3
Score: 91.3
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

ARC-C25 (Show measurement, test conditions, and source)

Source value: 91.3
Score: 91.3
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

TriviaQA (Show measurement, test conditions, and source)

Source value: 82.2
Score: 82.2
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

Natural Questions (Show measurement, test conditions, and source)

Source value: 40.1
Score: 40.1
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 1 benchmark row

Source Mistral AI release post

ARC-C (French) (Show measurement, test conditions, and source)

Source value: 59.1
Score: 59.1
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 2 French row

Source Mistral AI release post

HellaSwag (French) (Show measurement, test conditions, and source)

Source value: 79.2
Score: 79.2
Metric: percent
Unit: %
Direction: Higher is better
Source type: vendor-reported
Evaluator: Mistral AI release post
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Instruct model; Figure 2 French row

Source Mistral AI release post

Recorded measurement details

MedQA: Overall (Show measurement, test conditions, and source)

Source value: 62.139
Score: 62.139
Metric: accuracy
Unit: %
Benchmark version: 1
Category: health
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: vals-medqa:1:overall
Source type: independent-evaluator
Evaluator: Vals AI
Status: active
Retrieved at: 2026-09-29
Methodology: Source model ID: together/mistralai/Mixtral-8x22B-Instruct-v0.1. Source model name: Mixtral 8x22B. Evaluator: Vals AI. Dataset type: public. Leaderboard date: 2026-04-16. Latency in seconds: 8.946. Cost per task in USD: 0.000918. Source ID: vals-ai. Evidence capture ID: vals-medqa-2026-09-29-2886a92806d7. Source payload hash: 2886a92806d7b2e892c96203b7c7b462a4e28ec54dae900401734c38aa443883. Parser version: vals-astro-v2. Imported at: 2026-09-29

Source Vals AI

Page 1 of 3

Sources and data date

Type	Evidence and data date
Price	Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. Source: Pricing · API pricing
Input	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Mixtral 8x22B model card · Editorial description reviewed October 2, 2026
Output	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Mixtral 8x22B model card · Editorial description reviewed October 2, 2026
Research status	Research date September 8, 2026. 3 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not found in the inspected sources	knowledge-cutoff Original wording Inventory has February 2024, but checked official card/release does not publish an exact cutoff.
Not found in the inspected sources	training-tokens Original wording No token count found.
Not found in the inspected sources	max-output-tokens Original wording No model-specific maximum output found.
Unresolved	AA benchmark rows Original wording No exact Artificial Analysis model page located in the checked public sitemap.
Additional source	Context window data
Additional source	Vals AI (retrieved September 29, 2026)
Additional source	Mistral AI release post (retrieved September 8, 2026) · Mixtral 8x22B announcement · Editorial description reviewed October 2, 2026
Additional source	Mistral model overview · Editorial description reviewed October 2, 2026
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Creative Commons Attribution 4.0

Mixtral 8x22B

Specifications and access

Published benchmarks

MedQA

Measurements without matching peer values

Recorded measurement details

Our tests

SVG short test

Mixtral 8x22B

Pelican on a bicycle

Popularity

Published prices

Sources and data date