What are the hardware requirements for Mistral Passthrough 8L 10B?

Mistral Passthrough 8L 10B requires approximately 14.5 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mistral Passthrough 8L 10B and how large is it?

Mistral Passthrough 8L 10B is developed by DeepKarkhanis, a model with 10b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Mistral Passthrough 8L 10B perform on standard benchmarks?

Mistral Passthrough 8L 10B has the following published scores: MMLU 64.06. Compare against reference models on this page or on the LLM Explorer leaderboards.

Mistral Passthrough 8L 10B by DeepKarkhanis — VRAM 14.5GB, 32K context

Name: Mistral Passthrough 8L 10B
Author: DeepKarkhanis

Mistral Passthrough 8L 10B is an open-source language model by DeepKarkhanis. Features: 10b LLM, VRAM: 14.5GB, Context: 32K, License: apache-2.0, Instruction-Based, Merged, LLM Explorer Score: 0.11, Arc: 67.6, HellaSwag: 86.2, MMLU: 64.1, GSM8K: 68.2.

Merged Model Endpoints compatible Instruct Mistral Mistralai/mistral-7b-instruct-... Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/DeepKarkhanis/Mistral-Passthrough-8L-10B

Mistral Passthrough 8L 10B Parameters and Internals

Model Type

text generation

Additional Notes

This model is a merged model using LazyMergekit from mistralai/Mistral-7B-Instruct-v0.2 twice.

Training Details

Methodology:

Merge of layers using LazyMergekit

Input Output

Accepted Modalities:

text

Output Format:

generated text

LLM Name	Mistral Passthrough 8L 10B
Repository 🤗	https://huggingface.co/DeepKarkhanis/Mistral-Passthrough-8L-10B
Merged Model	Yes
Model Size	10b
Required VRAM	14.5 GB
Updated	2026-08-24
Maintainer	DeepKarkhanis
Model Type	mistral
Instruction-Based	Yes
Model Files	1.9 GB: 1-of-8 2.0 GB: 2-of-8 2.0 GB: 3-of-8 1.9 GB: 4-of-8 2.0 GB: 5-of-8 2.0 GB: 6-of-8 1.9 GB: 7-of-8 0.8 GB: 8-of-8
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Mistral Passthrough 8L 10B 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 Passthrough 8L 10B

Best Alternatives	Context / RAM	Downloads	Likes
Occiglot10b DPO	32K / 19.7 GB	4	1

Note: green Score (e.g. "73.2") means that the model is better than DeepKarkhanis/Mistral-Passthrough-8L-10B.