# Mixtral-8x22B-Instruct-v0.1 [Canonical model page](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1) · [Model library](https://headlessthrones.com/models) Public catalog reference from Headless Thrones. Exact catalog ID: `mistral-ai-mixtral-8x22b-instruct-v0-1`. Catalog source-check date: 2026-10-23. This brief contains the public catalog answers and any catalog guide, not private drafts, historical editorial revisions or independently measured rankings. The canonical HTML page may also display separately published editorial. The short answer ## Context and architecture Context window: 64,000 tokens Documented context [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) Attention heads: 48 num\_attention\_heads Key/value heads: 8 num\_key\_value\_heads Hidden layers: 56 num\_hidden\_layers Attention-head and layer counts come from the [Pinned configuration](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/config.json). The checkpoint separately records `max_position_embeddings=65,536`; this is not an independent API allowance or tested usable context. The same file explicitly sets `sliding_window` to `null`; this does not establish unlimited context. The catalog marks the publisher's API listing as retired; downloadable weights are a separate access route. [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) Catalog record checked October 23, 2026. [Context details](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-context) · [Checkpoint and sources](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-repository) Current catalog answers ## What to know about Mixtral-8x22B-Instruct-v0.1 AI-compiled from linked model documentation. This is a catalog reference, not an independent performance review. Catalog checked October 23, 2026. [Exact model](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-identity) · [Access](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-access) · [Token limits](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-context) · [Dates & knowledge](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-dates) · [License](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-license) · [Weight details](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-instruct-v0-1#model-answer-repository) ### Which Mixtral-8x22B-Instruct-v0.1 does this page describe? Mixtral-8x22B-Instruct-v0.1 is a model record from **Mistral AI** in the Mixtral family. [Source](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1) The recorded model identifier or checkpoint is `v0.1`. Use that exact identifier when checking documentation or a provider listing. [Source](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/README.md) ### How can you access Mixtral-8x22B-Instruct-v0.1? The catalog records the historical **API and Open weights** access route for Mixtral-8x22B-Instruct-v0.1. [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) As of the catalog check, this version was marked retired. [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) Check the linked lifecycle information before starting a new integration. For a new integration, use the publisher’s replacement guidance and check its current account, region and endpoint requirements. Historical API access does not mean the retired endpoint can still be called. Open weights means the model weights are distributed for use under their stated terms. It does not establish a hosted API, an open-source license, or a hardware requirement. This record documents API access separately. ### What are Mixtral-8x22B-Instruct-v0.1's context and output limits? The documented **context window is 64,000 tokens**. [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) A separate maximum-output value is not established in this record. The context or input figure should not be used as an output allowance. Context, input and output limits describe different constraints. Do not add the figures together to infer a larger window. Check the endpoint's accounting for conversation history, system instructions, tools, media and generated tokens; a documented maximum is not a guarantee of answer quality or a measured usable capacity. ### What dates are documented for Mixtral-8x22B-Instruct-v0.1? **Release date: April 17, 2024.** [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) This is a release milestone, not a knowledge or training cutoff. The source labels this “Release date on Mistral model specification.” Neither a knowledge cutoff nor a training-data cutoff is established here. A release date, repository upload or filename cannot supply those missing dates. ### What license is recorded for Mixtral-8x22B-Instruct-v0.1? The recorded license label is **Apache 2.0**. [Source](https://docs.mistral.ai/models/mixtral-8x22b-0-1-0-3) Read the linked license and associated use terms for this exact version before redistribution, commercial deployment or fine-tuning. A label alone does not establish every permission, restriction or obligation, and it does not determine the terms of a separate hosted service. ### What is documented about the Mixtral-8x22B-Instruct-v0.1 checkpoint? This evidence refers to `mistralai/Mixtral-8x22B-Instruct-v0.1` at recorded revision `cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d`, checked October 23, 2026. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) Safetensors metadata records **140,630,071,296 stored parameters**. This is a stored-weight count, not necessarily the number activated per token, a download size or a GPU-memory requirement. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) Recorded architecture class: `MixtralForCausalLM`. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) Listed library: `vllm`. Check that library's version and model support before loading the weights. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) Repository language labels: en, es, it, de, and fr. These labels are not independent evaluations of language quality. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) Repository access gate: `false`. This records the repository's gate setting when checked; licensing and use terms still apply. [Repository metadata](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) #### What the configuration file says The pinned `config.json` contains these values. [Pinned configuration](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/config.json) Configuration verified 2026-10-26. `max_position_embeddings`: 65,536 `sliding_window`: null `torch_dtype`: bfloat16 `num_attention_heads`: 48 `num_key_value_heads`: 8 `num_hidden_layers`: 56 `hidden_size`: 6,144 Configuration values describe the checkpoint and its implementation. In particular, max\_position\_embeddings, sliding\_window and rope\_scaling do not independently establish an API context limit, tested usable length or hardware requirement. Runtime, quantization, cache, batch size and input length affect deployment behavior. The explicit null is a recorded configuration value, not missing evidence. It does not establish unlimited context. Read the retrieved model card for setup and qualifications. [Model card](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/README.md) **Catalog qualifications:** Official publisher model repository; listed status confirms a catalog entry, not a verified hosted API service. Lifecycle status refers to the Mistral API; publicly listed weights may remain usable under their license. Mistral context shorthand 64k is represented in nominal decimal tokens. [Source 1](https://huggingface.co/api/models/mistralai/Mixtral-8x22B-Instruct-v0.1) · [Source 2](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/README.md) · [Source 3](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/resolve/cc88a6cc19fbd17d9f1c0ee0b0d70a748dce698d/config.json) **Evidence still missing here:** complete input/output pricing, knowledge and training-data cutoffs, and tested hardware requirements. Missing documentation is not evidence that a feature is unavailable. Headless Thrones has not independently benchmarked this model for this reference. Evaluate it on representative work before choosing it; these facts do not rank its accuracy, reliability, speed or suitability against another model. Keep the exact version in view ## Other Mixtral models from Mistral AI. These are separate records in the same publisher family. A shared family name does not establish equal capabilities or interchangeable licenses. ### [Mixtral-8x7B-Instruct-v0.1](https://headlessthrones.com/models/mistral-ai-mixtral-8x7b-instruct-v0-1) API · Open weights · 32,000 context tokens ### [Mixtral-8x22B-v0.1](https://headlessthrones.com/models/mistral-ai-mixtral-8x22b-v0-1) Open weights ### [Mixtral-8x7B-v0.1](https://headlessthrones.com/models/mistral-ai-mixtral-8x7b-v0-1) Open weights