[whatmodel.sh](../../README.md) / Mixtral 8x22B Instruct v0.1
# Mixtral 8x22B Instruct v0.1
Mistral · released 2024-04 · 141B params (39B active) · apache-2.0
**score 49.9** ±0.6 · #372 of 732 · from 20 benchmark results across 15 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Mixtral 8x22B Instruct v0.1 among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁●▆█▄▅▅▄▄▂▁ #291 of 313* 0.8% 1
coding ▁▁▁▃▅▇●▇▅▄▃▁ #279 of 469 56% 8
chat & writing ▁▁▁▃▆●▆█▆▇▇▃ #296 of 402 29% 4
agent tasks · #291 of 313 · 0.8% of its score
Beats [Bielik 11B v2.3 Instruct](bielik-11b-v2-3-instruct.md) here, though they rank higher overall. Trails [Granite 4.0 H 350M](granite-4-0-h-350m.md), [Qwen3 0.6B](qwen3-0-6b.md), [Qwen2.5 1.5B Instruct](qwen2-5-1-5b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BFCL v3 41.8% ▁▁▁▁▁●▅█▅▄▁▁ beats 19/114 medium · avg 51% 0.8% blunt · at its level
coding · #279 of 469 · 56% of its score
Beats [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md), [GLM-4.7-Flash](glm-4-7-flash.md) here, though they rank higher overall. Trails [DeepSeek LLM 67B Chat](deepseek-llm-67b-chat.md), [Mistral NeMo Instruct (2407)](mistral-nemo-instruct-2407.md), [QwQ-32B-Preview](qwq-32b-preview.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HumanEval 76.2% ▁·▁▁▃▄▆▅●▃▁▁ beats 75/99 easy · avg 76% 19% sharp
HumanEval+ 72.0% ▁·▁▁▁▃▄▄▄●▃▁ beats 79/99 easy · avg 70% 17% sharp · at its level
HumanEvalSynthesize (HumanEvalPac… 64.8% ▁·▂▁▂▆▅█▄▄▁● beats 34/35 medium · avg 62% 5.4% at its level
HumanEvalExplain (HumanEvalPack, … 56.6% ▄▇█▃▄▇▄▁▇▃·● beats 35/35 medium · avg 46% 4.3% at its level
HumanEvalFix (HumanEvalPack, 6-la… 56.1% ▁·▁▁▄██▄█▁▅● beats 35/35 medium · avg 51% 4.2% sharp · at its level
MBPP+ 64.3% ▁···▁▁▂▅▇●▆▂ beats 47/75 easy · avg 66% 3.6% at its level
WebApp1K-React pass@1 (onekq-ai) 30.7% ▃▁▁●▁▄▅▃▄▇▁▄ beats 6/31 medium · avg 40% 1.5% sharp · at its level · noisy
SciCode main problems (official l… 0.0% ●···█··▂·▃·▂ beats 1/17 very hard · avg 1.0% 0.8% sharp · too hard for it
chat & writing · #296 of 402 · 29% of its score
Beats [Ministral 3 14B Reasoning 2512](ministral-3-14b-reasoning.md), [Gemma 4 E4B](gemma-4-e4b.md), [OpenThinker3 7B](openthinker3-7b.md) here, though they rank higher overall. Trails [EXAONE 3.5 2.4B Instruct](exaone-3-5-2-4b.md), [LFM2 700M](lfm2-700m.md), [Flan-T5 XXL](flan-t5-xxl.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AlpacaEval 2 length-controlled 30.9% ▁▂▂▃▆▆█●▄▄▁▁ beats 80/122 medium · avg 35% 14% sharp · at its level
Arena-Hard (v0.1, GPT-4-1106 judg… 36.4% ▂▁▁▅█●▄▃▃▅▃▁ beats 31/60 medium · avg 47% 8.5% sharp · at its level
LMArena (text) 1229.7 ▁▁▁▂▃●▃▆▆▇█▃ beats 82/385 avg 1260 3.4% blunt
Polyrating rating (bias-aware re-… 1152.0 ▂▁▂▁▃▇█▆●▄▂▂ beats 67/95 avg 1187 3.1% typical
other tests (7)
test result vs the others in its run difficulty counts why
BigCodeBench (Instruct) 40.4% ▃▂·▆▆██▆●▃▃▃ beats 25/33 medium · avg 39% 4.0% at its level
BigCodeBench (Complete) 49.6% ▂▁▁▄▂▅▇█▆●▂▂ beats 50/58 medium · avg 48% 3.8% at its level
CPTUB average (SpeakLeash, 0-5 ju… 71.2% ▁▂▂▂▃█▃●▅▄▂▃ beats 57/95 medium · avg 65% 2.5% at its level
Polish EQ-Bench v2 (SpeakLeash) 67.6% ▁··▁▁▂▂▆▅▆●▆ beats 46/70 medium · avg 62% 2.0% at its level
Open PT-LLM Leaderboard average (… 76.6% ▃▁▁▁▁▁▁▁▂█●▁ beats 235/276 medium · avg 72% 0.8% at its level · noisy
RULER 64k 84.7% ▁▁·▁·▂▃▄●▂·▁ beats 13/22 easy · avg 82% 0.7% typical
Vectara hallucination leaderboard… 95.3% ▁▁▂▃▄▄●▆▃▂▂▁ beats 66/140 very easy · avg 94% 0.2% blunt · maxed out · noisy
source results share of the score
EvalPlus / EvalPerf published evaluatio… 3 40% ▇▇▇▇▇
Granite Code Models: A Family of Open F… 3 14% ▇▇
AlpacaEval 1.0 and 2.0 leaderboards 1 14% ▇▇
Arena-Hard-Auto v0.1 leaderboard 1 9% ▇
BigCodeBench paper 2 8% ▇
LMArena leaderboard dataset 1 3% ▇
Polyrating: A Cost-Effective and Bias-A… 1 3% ▇
SpeakLeash CPTUB 1 3% ▇
SpeakLeash Polish EQ-Bench leaderboard 1 2% ▇
WebApp1K models leaderboard 1 2% ▇
Open Portuguese LLM Leaderboard 1 1% ▇
SciCode leaderboard 1 1% ▇
Berkeley Function Calling Leaderboard r… 1 1% ▇
NVIDIA RULER repository README results … 1 1% ▇
Vectara hallucination leaderboard 1 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 49.9
Licences: EvalPlus / EvalPerf published evaluation runs ([Apache-2.0](../../licenses/evalplus-published-runs-Apache-2.0.txt)) · Granite Code Models: A Family of Open Foundation Models for Code Intelligence ([CC BY 4.0](../../licenses/paper-2405.04324v1-CC-BY-4.0.txt)) · AlpacaEval 1.0 and 2.0 leaderboards ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · Arena-Hard-Auto v0.1 leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-v0-1-Apache-2.0.txt)) · BigCodeBench paper ([CC BY 4.0](http://creativecommons.org/licenses/by/4.0/)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · SpeakLeash CPTUB ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · SpeakLeash Polish EQ-Bench leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · WebApp1K models leaderboard ([MIT](../../licenses/webapp1k-leaderboard-MIT.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · SciCode leaderboard ([Apache-2.0](../../licenses/scicode-Apache-2.0.txt)) · Berkeley Function Calling Leaderboard result archive ([Apache-2.0](../../licenses/bfcl-result-archive-v4-Apache-2.0.txt)) · NVIDIA RULER repository README results table ([Apache-2.0](https://github.com/NVIDIA/RULER/blob/c3f5e3b4f87f97e048793bb510a3a6b19a46bf3a/LICENSE)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Mixtral 8x22B Instruct v0.1 (Q4_K_M, needs ~85.6 GB, from [huggingface.co/MaziyarPanahi](https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF/tree/9f2f6c5ec37f9bce5f5f3a7ff07b11d573443e62))
```sh
llama cli -hf MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF:Q4_K_M
```
Other versions: Q8_0 (~149.4 GB) `llama cli -hf MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF:Q8_0` · Q2_K (~52.1 GB) `llama cli -hf MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF:Q2_K`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Mixtral 8x22B Instruct v0.1 (Q4_K_M, needs ~85.6 GB, from [huggingface.co/MaziyarPanahi](https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF/tree/9f2f6c5ec37f9bce5f5f3a7ff07b11d573443e62))
```sh
lms get https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~149.4 GB) `lms get https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF@Q8_0 -y` · Q2_K (~52.1 GB) `lms get https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF@Q2_K -y` · 4bit (~79.1 GB, Mac only) `lms get https://huggingface.co/mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit -y` · 8bit (~149.6 GB, Mac only) `lms get https://huggingface.co/mlx-community/Mixtral-8x22B-Instruct-v0.1-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Mixtral 8x22B Instruct v0.1 (Q4_K_M, needs ~85.1 GB, from [ollama.com](https://ollama.com/library/mixtral/tags))
```sh
ollama run mixtral:8x22b-instruct-v0.1-q4_K_M
```
Other versions: Q8_0 (~169.7 GB) `ollama run mixtral:8x22b-instruct-v0.1-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Mixtral 8x22B Instruct v0.1 (4bit, needs ~79.1 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit/tree/0ec04c0321fbcde42ac53fe51587d75d7328ba85))
```sh
mlx_lm.chat --model mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit
```
Other versions: 8bit (~149.6 GB, Mac only) `mlx_lm.chat --model mlx-community/Mixtral-8x22B-Instruct-v0.1-8bit`
Model repository and documentation: [huggingface.co/mistralai](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1)
[interactive version](https://whatmodel.sh/model/mixtral-8x22b-instruct-v0-1) · [JSON](https://whatmodel.sh/api?format=json&all=1)