Mixtral 8x22B Instruct
Mistral AI · released Apr 16, 2024 · mistralai/Mixtral-8x22B-Instruct-v0.1
- Type
- Open weightsApache License 2.0
- Params
- 141B
- Context
- 66K
39B active per word · about 49K words of context
Our take
Written Aug 3, 2026Mixtral is a large downloadable text model from Mistral AI that uses a mixture-of-experts design, keeping 39 billion parameters active per word out of 140.6 billion total. Released in 2024 under a permissive Apache licence, it handles document-scale requests up to 65,536 tokens but scores near the bottom of the leaderboards we track for knowledge and reasoning.
Pick this when you need a permissive licence for self-hosting or redistribution, or for long-context document work where 65,536 tokens is enough. Use it if you want identical pricing across the two providers that carry it, with no arbitrage hunting needed. Skip it if you need strong graduate-level reasoning, competitive instruction-following scores, or any price competition between hosts.
The case for it
- Apache 2.0 licence allows commercial use, fine-tuning and redistribution without restriction.
- 65,536-token request limit supports document-scale text work.
- Identical pricing across both tracked providers removes arbitrage friction.
The case against it
- Academic benchmark scores lag on knowledge and reasoning: MMLU-Pro at 38.7% and GPQA Diamond at 16.4%, both near or below random-guess territory on graduate-level material.
- Arena Elo scores place it in the lower tier of rated instruction-following models, with its lowest category score in instruction following itself.
- No price competition between its only two hosts, both listing the same rate.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)139th of 143 · 1229
CodingWriting and fixing code on its own
Arena Coding138th of 143 · 1276.9
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Mixtral 8x22B Instruct for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Mixtral 8x22B Instruct placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Ultra (64-core GPU) · 128 GB
Borderline fit on an estimated size. It leaves 3 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 2 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $2.00 in / $6.00 out
- Context served
- 66K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $2.00 / $6.00 | 66K | not measured | Unknown | Unknown | Unknown |
| Mistral AI | $2.00 / $6.00 | 66K | 97 tok/s | No | Yes30 days | Unknown |
Mistral AIAcross the 2 listings we hold: 1 say they do not train on prompts, 0 say they do and 1 do not say. 0 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Mistral AI | ✓ | ✓ | ✓ |
Tool calling: 2 of 2 listings say yes. JSON output: 2 of 2 listings say yes. Strict schema: 2 of 2 listings say yes.
When we formed this view
Dates behind this page
Prices last checked 4d ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 2 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- mistralai/Mixtral-8x22B-Instruct-v0.1
- Architecture
- Mixture of experts
- Modality record
- text+file->text
- Catalogue slug
- mistralai-mixtral-8x22b-instruct