What are the hardware requirements for Mistral 22B V0.1?

Mistral 22B V0.1 requires approximately 44.7 GB of VRAM and supports a context window of 64K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mistral 22B V0.1 and how large is it?

Mistral 22B V0.1 is developed by Vezora, a model with 22b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Mistral 22B V0.1 perform on standard benchmarks?

Mistral 22B V0.1 has the following published scores: MMLU 48.55. Compare against reference models on this page or on the LLM Explorer leaderboards.

Mistral 22B V0.1 by Vezora — VRAM 44.7GB, 64K context

Name: Mistral 22B V0.1
Author: Vezora

Mistral 22B V0.1 is an open-source language model by Vezora. Features: 22b LLM, VRAM: 44.7GB, Context: 64K, License: apache-2.0, HF Score: 49.8, LLM Explorer Score: 0.16, Arc: 49.2, HellaSwag: 72.8, MMLU: 48.6, TruthfulQA: 47.3, WinoGrande: 74.5, GSM8K: 6.4.

Conversational Endpoints compatible Mistral Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Vezora/Mistral-22B-v0.1

Mistral 22B V0.1 Benchmarks

ARC: 49.15 vs 96.7 (so35)^-49.2%

HellaSwag: 72.81 vs 95.3 (gpt4)^-23.6%

MMLU: 48.55 vs 88.3 (so35)^-45%

TruthfulQA: 47.31 vs 59 (gpt4)^-19.8%

WinoGrande: 74.51 vs 87.5 (gpt4)^-14.8%

GSM8K: 6.44 vs 96.4 (so35)^-93.3%

LLME Score: 0.16428

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Mistral 22B V0.1 (Vezora/Mistral-22B-v0.1)

🌟 Advertise your project 🚀

Mistral 22B V0.1 Parameters and Internals

Model Type

Dense Model

Use Cases

Areas:

Experimentation, Mathematics

Applications:

Proficiency in mathematical tasks

Limitations:

Experimental model, Trained with limited data and time.

Additional Notes

Stay tuned for the release of V.2 with enhanced features.

Training Details

Data Sources:

500 regular human written Q/A, 500 tested python examples

Data Volume:

Small scale due to experimental model

Methodology:

First MOE to dense model conversion

Training Time:

Trained in less than an hour

Model Architecture:

22B parameter dense model

Release Notes

Version:

V.01

Date:

April 11

Notes:

First MOE to dense model conversion; trained on a limited dataset in a short time, expected performance comparable to Llama 1.

LLM Name	Mistral 22B V0.1
Repository 🤗	https://huggingface.co/Vezora/Mistral-22B-v0.1
Model Size	22b
Required VRAM	44.7 GB
Updated	2026-04-25
Maintainer	Vezora
Model Type	mistral
Model Files	4.9 GB: 1-of-9 5.0 GB: 2-of-9 5.0 GB: 3-of-9 4.9 GB: 4-of-9 5.0 GB: 5-of-9 5.0 GB: 6-of-9 4.9 GB: 7-of-9 5.0 GB: 8-of-9 5.0 GB: 9-of-9
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	65536
Model Max Length	65536
Transformers Version	4.39.3
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	bfloat16

Quantized Models of the Mistral 22B V0.1

Model	Likes	Downloads	VRAM
Mistral 22B V0.1 AWQ	1	9	12 GB

Best Alternatives to Mistral 22B V0.1

Best Alternatives	Context / RAM	Downloads	Likes
MS Schisandra 22B V0.2	128K / 44.7 GB	17	9
...ntheon RP Pure 1.6.2 22B Small	128K / 44.7 GB	14	33
MS Meadowlark 22B	128K / 44.7 GB	11	16
...rker The Final Abomination 22B	128K / 44.7 GB	6	6
...er The Final Transgression 22B	128K / 44.7 GB	3	3
...Darker The Final Directive 22B	128K / 44.7 GB	6	0
Retrograde Omega M 22B V1.0	128K / 44.7 GB	5	0
The Omega Directive M 22B V1.0	128K / 44.7 GB	12	2
MS Inky 2409 22B	128K / 44.7 GB	6	0
MS Quadrosiac 2409 22B	128K / 44.7 GB	9	0

Note: green Score (e.g. "73.2") means that the model is better than Vezora/Mistral-22B-v0.1.

Rank the Mistral 22B V0.1 Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 53254 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Mistral 22B V0.1 by Vezora