Model Library
MistralReleased April 17, 2024Mixtral 8x22B
Mistral · Released April 17, 2024
Intelligence score
4
At a glance
- Intelligence
- 4
- Speed
- —
- Latency
- —
- Input price
- Not published
- Output price
- Not published
- Context window
- 65Ktokens
- Parameters
- 141billions
Summary
Comparison summary
Mixtral 8x22B Instruct scores 4 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 7).
Specifications
Technical specifications
| Model name | Mixtral 8x22B |
|---|---|
| Provider | Mistral |
| Release date | April 17, 2024 |
| Intelligence score | 4 |
| Speed | — |
| Latency (TTFT) | — |
| Input price | Not published |
| Output price | Not published |
| Context window | 65,000 tokens |
| Parameters | 141B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: mistral-8x22b-instruct |
FAQ · 16
Frequently asked questions
Mixtral 8x22B Instruct was released on April 17, 2024.
Mixtral 8x22B Instruct was created by Mistral.
Mixtral 8x22B Instruct scores 4 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 7).
No, Mixtral 8x22B Instruct is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Mixtral 8x22B Instruct supports text only input.
Mixtral 8x22B Instruct supports text only output.
No, Mixtral 8x22B Instruct does not support image input. It can only process text.
No, Mixtral 8x22B Instruct is not multimodal. It only supports text only input.
Mixtral 8x22B Instruct has a context window of 65k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Mixtral 8x22B Instruct is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Mixtral 8x22B Instruct has 141 billion parameters (39 billion active).
Mixtral 8x22B Instruct is a Mixture of Experts (MoE) model with 141 billion total parameters, but only 39 billion active parameters are used during inference.
Mixtral 8x22B Instruct is released under the Apache 2.0 license. This license allows commercial use.
Mixtral 8x22B Instruct achieves a score of 4 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Mixtral 8x22B Instruct is an open weights model that can be self-hosted.
Mixtral 8x22B Instruct is an open weights model that can be downloaded and self-hosted.