whatmodel.sh / Mixtral 8x22B Instruct v0.1

Mixtral 8x22B Instruct v0.1

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Mixtral 8x22B Instruct v0.1 is an open-weights language model from Mistral, released April 2024. Its published size is 141B parameters (39B active per token) and a 64k-token context window. Licence: apache-2.0.

Catalogue identity notes

Arena entry 'mixtral-8x22b-instruct-v0.1' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'mixtral-8x22b-instruct-v0.1' as 'Mixtral-8x22b-Instruct-v0.1' (Mistral, Apache 2.0), linking https://mistral.ai/news/mixtral-8x22b/. Maker repo mistralai/Mixtral-8x22B-Instruct-v0.1 @cc88a6c: licence tag apache-2.0. Parameters: mistral.ai/news/mixtral-8x22b: 'uses only 39B active parameters out of 141B'. Released 2024-04: month the maker's Hugging Face repo was created (createdAt 2024-04-16).

48.6±0.6#369 of 719 LLMsfrom 20 benchmark results across 15 sources

$ how it compares

The LLMs ranked either side of Mixtral 8x22B Instruct v0.1, out of 828 with a score, then more from Mistral.

#modelscorewhereRAMreleased
440Claude 3 Sonnet Anthropic49±1comparison only2024-02
441AceReason-Nemotron-1.1-7B NVIDIA49±2local~6GB2025-06
442Mixtral 8x22B Instruct v0.1 Mistral49±1local~56GB2024-04
443Yi-Coder 9B Chat 01.AI49±1local~6.8GB2024-08
444Qwen2-Math 7B Instruct Alibaba49±2local~6GB2024-08
more from Mistral
79Mistral Large 4 (public preview) Mistral71±2api $2.06/M2026-10
174Mistral Medium 3.5 (128B) Mistral65±1api $3.00/M +local~42GB2026-04
180Mistral Medium 3.1 (2508) Mistral64±2comparison only2025-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #278 of 300 * · 0.7% of its score · 1 test

Beats Bielik 11B v2.3 Instruct here, though they rank higher overall. Trails Granite 4.0 H 350M, Qwen3 0.6B, Qwen2.5 1.5B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL v341.8% ✎▁▁▁▁▁▃▅█▅▄▁▁ beats 19/114medium · avg 52%▇ 0.7%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #270 of 458 · 59% of its score · 8 tests

Beats Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B), GLM-4.7-Flash here, though they rank higher overall. Trails DeepSeek LLM 67B Chat, Mistral NeMo Instruct (2407), QwQ-32B-Preview here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval76.2% ✎▁·▁▁▃▄▆▅█▃▁▁ beats 75/99easy · avg 79%▇▇▇▇▇▇▇▇▇ 22%sharp
HumanEval+72.0% ✎▁·▁▁▁▃▄▄▄█▃▁ beats 79/99easy · avg 74%▇▇▇▇▇▇▇ 18%sharp · at its level
HumanEvalSynthesize (HumanEvalPack, 6-language average)64.8% ✎▁·▂▁▂▆▅█▄▄▁▃ beats 34/35medium · avg 64%▇▇ 5.4%at its level
HumanEvalExplain (HumanEvalPack, 6-language average)56.6% ✎▄▇█▃▄▇▄▁▇▃·▃ beats 35/35medium · avg 49%▇▇ 4.4%at its level
HumanEvalFix (HumanEvalPack, 6-language average)56.1% ✎▁·▁▁▄██▄█▁▅▅ beats 35/35medium · avg 56%▇▇ 4.2%sharp · at its level
MBPP+64.3% ✎▁···▁▁▂▅▇█▆▂ beats 47/75easy · avg 68%▇ 3.5%at its level
WebApp1K-React pass@1 (onekq-ai)30.7% ✎▃▁▁█▁▄▅▃▄▇▁▄ beats 6/31medium · avg 44%▇ 1.4%sharp · at its level · noisy
SciCode main problems (official leaderboard)0.0% ✎▃···█··▂·▃·▂ beats 1/17very hard · avg 1.1%▇ 0.7%sharp · too hard for it
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #288 of 393 · 27% of its score · 4 tests

Beats Ministral 3 14B Reasoning 2512, Gemma 4 E4B, OpenThinker3 7B here, though they rank higher overall. Trails Flan-T5 XXL, LFM2 700M, EXAONE 3.5 2.4B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AlpacaEval 2 length-controlled30.9% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 80/122medium · avg 39%▇▇▇▇▇ 12%at its level
Arena-Hard (v0.1, GPT-4-1106 judge)36.4% ✎▂▁▁▅█▅▄▃▃▅▃▁ beats 31/60medium · avg 52%▇▇▇ 8.5%sharp · at its level
Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024)1152.0 ✎▂▁▂▁▃▇█▆▇▄▂▂ beats 67/95avg 1201▇ 3.1%blunt
LMArena (text)1229.5 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 83/379avg 1268▇ 3.0%blunt
other tests7 not in a task · 13% of its score
testresultvs the others in its rundifficultycountswhy
BigCodeBench (Instruct)40.4% ✎▃▂·▆▆██▆▅▃▃▃ beats 25/33medium · avg 41%▇▇ 3.8%at its level
BigCodeBench (Complete)49.6% ✎▂▁▁▄▂▅▇█▆▃▂▂ beats 50/58medium · avg 50%▇ 3.7%at its level
CPTUB average (SpeakLeash, 0-5 judge scale)71.2% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 57/95easy · avg 66%▇ 2.1%at its level
Polish EQ-Bench v2 (SpeakLeash)67.6% ✎▁··▁▁▂▂▆▅▆█▆ beats 46/70medium · avg 64%▇ 1.7%at its level
RULER 64k84.7% ✎▁▁·▁·▂▃▄█▂·▁ beats 13/22easy · avg 83%▇ 0.7%typical
Open PT-LLM Leaderboard average (9 Portuguese tasks)76.6% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 235/276medium · avg 73%▇ 0.7%at its level · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)95.3% ✎▁▁▂▃▄▄█▆▃▂▂▁ beats 66/140very easy · avg 94%▇ 0.2%blunt · maxed out · noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Mixtral 8x22B Instruct v0.1 · needs ~85.6 GB · from huggingface.co/MaziyarPanahi
    chat in the terminal
    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)
    llama serve -hf MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Mixtral 8x22B Instruct v0.1 · needs ~85.6 GB · from huggingface.co/MaziyarPanahi

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-Instruct-v0.1-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Mixtral 8x22B Instruct v0.1 · needs ~85.1 GB · from ollama.com
    chat in the terminal
    ollama run mixtral:8x22b-instruct-v0.1-q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Mixtral 8x22B Instruct v0.1 · needs ~79.1 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Mixtral-8x22B-Instruct-v0.1-4bit

Weights and documentation: huggingface.co/mistralai ↗

$ how the score was worked out · EvalPlus / EvalPerf published… 43%, Granite Code Models: A Family… 14%, AlpacaEval 1.0 and 2.0 leader… 12% and 12 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EvalPlus / EvalPerf published evaluation runs3▇▇▇▇▇▇▇▇▇▇▇▇▇ 43%Apache-2.0
Granite Code Models: A Family of Open Foundation Models for Code Intelligence — authors' own runs3▇▇▇▇ 14%CC BY 4.0
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)1▇▇▇▇ 12%Apache-2.0
Arena-Hard-Auto v0.1 leaderboard (repository README, pre-v2)1▇▇▇ 8%Apache-2.0
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v42▇▇ 8%CC BY 4.0
Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3)1▇ 3%CC-BY-4.0
LMArena leaderboard dataset1▇ 3%CC BY 4.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇ 2%MIT
SpeakLeash Polish EQ-Bench leaderboard1▇ 2%Apache-2.0
WebApp1K models leaderboard (onekq-ai)1▇ 1%MIT
SciCode leaderboard (repository README)1▇ 1%Apache-2.0
NVIDIA RULER repository README results table1▇ 1%Apache-2.0
Berkeley Function Calling Leaderboard result archive1▇ 1%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 1%Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 0%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 48.6

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json