What are the hardware requirements for Mistral 22B V0.2 GGUF?

Mistral 22B V0.2 GGUF requires approximately 8.3 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 V0.2 GGUF and how large is it?

Mistral 22B V0.2 GGUF is developed by failspy, 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 V0.2 GGUF perform on standard benchmarks?

Mistral 22B V0.2 GGUF has the following published scores: MMLU 54.63. Compare against reference models on this page or on the LLM Explorer leaderboards.

Mistral 22B V0.2 GGUF by failspy — VRAM 8.3GB, 64K context

Name: Mistral 22B V0.2 GGUF
Author: failspy

Mistral 22B V0.2 GGUF is an open-source language model by failspy. Features: 22b LLM, VRAM: 8.3GB, Context: 64K, License: apache-2.0, Quantized, LLM Explorer Score: 0.14, Arc: 52.4, HellaSwag: 78.6, MMLU: 54.6, GSM8K: 27.

Endpoints compatible F16 Ggml Gguf Mistral Q2 Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/failspy/Mistral-22B-v0.2-GGUF

Mistral 22B V0.2 GGUF Parameters and Internals

Model Type

Dense, Not MOE

Additional Notes

The model exhibits strong mathematical abilities and improved coding skills. It has agent abilities and can perform multi-turn conversations. It has a 32k sequence length.

Training Details

Data Volume:

8x more data than v0.1

Methodology:

Knowledge distilled from all experts into a single dense model

Context Length:

32000

Input Output

Input Format:

Guanaco chat template

Release Notes

Version:

v0.2

Date:

April 13

Notes:

This model is not a single trained expert, instead it's a compressed MOE model, turning it into a dense 22B model. It has been trained on 8x more data than v0.1.

LLM Name	Mistral 22B V0.2 GGUF
Repository 🤗	https://huggingface.co/failspy/Mistral-22B-v0.2-GGUF
Base Model(s)	Vezora/Mistral-22B-v0.2 Vezora/Mistral-22B-v0.2
Model Size	22b
Required VRAM	8.3 GB
Updated	2026-08-01
Maintainer	failspy
Model Type	mistral
Model Files	8.3 GB 10.8 GB 11.7 GB 9.6 GB 13.3 GB 12.7 GB 15.7 GB 15.3 GB 18.2 GB 23.6 GB 44.5 GB
GGML Quantization	Yes
GGUF Quantization	Yes
Quantization Type	gguf\|ggml\|q2\|q4_k\|q5_k
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	65536
Model Max Length	65536
Transformers Version	4.40.0.dev0
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Mistral 22B V0.2 GGUF capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to Mistral 22B V0.2 GGUF

Best Alternatives	Context / RAM	Downloads	Likes
Codestral 22B V0.1 GGUF	32K / 8.3 GB	361	1
Codestral 22B V0.1 GGUF	32K / 8.3 GB	173	0
MwM 22B Instruct	128K / 44.7 GB	6	0
... Instruct 0.2 Chkpt 200 16 Bit	128K / 44.7 GB	6	1
...l 22B V0.1 Hf FIM Fix Bnb 4bit	32K / 13.1 GB	1	1
Text2cypher Codestral 16bit	32K / 44.7 GB	7	3
... Abliterated V3.8.0bpw H8 EXL2	32K / 21.2 GB	2	1
Codestral 22B V0.1 EXL2 8.0bpw	32K / 20.9 GB	9	3
Codestral 22B V0.1 EXL2 6.0bpw	32K / 16.9 GB	7	1
...estral 22B V0.1 Hf 2 2bpw EXL2	32K / 6.6 GB	3	1

Note: green Score (e.g. "73.2") means that the model is better than failspy/Mistral-22B-v0.2-GGUF.