Mixtral 8x7B Specs, Benchmarks & Downloads
Mixtral 8x7B
Sparse mixture of experts model
Status
active
Parameters
46.00
Context Window
32768
Modality
Text
Release Date
Dec 11, 2023
About
Explore Mixtral 8x7B, a text AI model developed by Mistral AI. View specifications, context window, architecture, license, supported platforms, features, benchmarks, official downloads, and related articles.
Mixtral 8x7B uses a mixture-of-experts architecture for efficient high performance.
Family
MixtralDeveloper
Mistral AIArchitecture
Mixture of ExpertsLicense
Mistral LicenseHardware Requirements
Minimum:
Use Case: Local inference
GPU Minimum: RTX 4090
GPU VRAM: 24GB
System RAM: 64GB
Storage: 100GB
Operating System: Linux
Mixtral 8x7B requires larger memory
Recommended:
Use Case: Multi GPU inference
GPU Minimum: 2x RTX 4090
GPU VRAM: 48GB
System RAM: 128GB
Storage: 150GB
Operating System: Linux
Recommended for MoE deployment
High Performance:
Use Case: Production serving
GPU Minimum: 4x A100
GPU VRAM: 320GB
System RAM: 256GB
Storage: 300GB
Operating System: Linux
Enterprise inference
Model Deployment Readiness
Model Ecosystem Support
21%
Platforms
42%
Features
1%
Downloads
20%
Platform Compatibility
42%
Supported across 5 of 12 platforms
Context Window
Industry-standard context window comparison
32,768
tokens
32K
4K
32K
128K
1M
2M
Database Rank
#30
Percentile
Top 71%
Capacity
High
Benchmark Performance
Evaluation profile across publicly reported benchmarks
|
HumanEval
|
Coding | 40.2 % | 2023 |
|
MMLU
|
Knowledge | 70.6 % | 2023 |
|
GPQA
|
Reasoning | 30.4 % | 2023 |
|
HumanEval
|
Coding | 40.2 % | 2023 |
|
GSM8K
|
Mathematics | 74.4 % | 2023 |
|
HellaSwag
|
Reasoning | 84.9 % | 2023 |
Highest Benchmark
HellaSwag
84.9%
Benchmarks Available
6
Latest Evaluation
Dec 2023
