Home/Models/Mixtral 8x22B Instruct
Mistral AI logo

Mixtral 8x22B Instruct

by Mistral AI · Released Apr 2024

Open Source
24.6
avg score
Rank #197
Compare
Better than 15% of all models
Context
66K tokens (~33 books)
Input $/1M
$2.00
Output $/1M
$6.00
Type
text
License
Open Source
Benchmarks
5 tested
Data updated today
About

Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding,...

Tested on 5 benchmarks with 23.5% average. Top scores: MMLU (70.4%), MATH level 5 (24.2%), GPQA diamond (12.1%).

Looking for similar performance at lower cost?
Gemma 3 27B scores 25.1 (102% as good) at $0.08/1M input · 96% cheaper
Capabilities
coding
5.3
#141 globally
math
24.2
#148 globally
knowledge
41.2
#147 globally
Benchmark Scores
Compare All
Tested on 5 benchmarks · Ranked across 3 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
Cybench

Capture-the-flag cybersecurity challenges. Tests vulnerability analysis, reverse engineering, cryptography, and exploitation skills.

7.5
WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

3.2
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

24.2
MMLU

Massive Multitask Language Understanding. 57 subjects from STEM, humanities, and social sciences. The most widely-cited knowledge benchmark.

70.4
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

12.1
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
Community
BenchGecko API
mixtral-8x22b-instruct
Specifications
  • Typetext
  • Context66K tokens (~33 books)
  • ReleasedApr 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.010
Available On
Mistral AI logoMistral AI$2.00
Share & Export
Tweet
Mixtral 8x22B Instruct is an open-source text AI model by Mistral AI, released in April 2024. It has an average benchmark score of 24.6. Context window: 66K tokens.