LLM EXPLORER 59,598 MODELS INDEXED

E5 Mistral 7B Instruct by intfloat

By intfloat · 449972 downloads
HF Repo ↗ Copy URL

E5 Mistral 7B Instruct is an open-source language model by intfloat. Features: 7b LLM, VRAM: 14.2GB, Context: 32K, License: mit, Instruction-Based, LLM Explorer Score: 0.25, Arc: 29.6, HellaSwag: 27.1, MMLU: 23.1.

  Arxiv:2104.08663   Arxiv:2210.07316   Arxiv:2212.03533   Arxiv:2401.00368   Deploy:azure   En   Endpoints compatible   Eval-results   Feature-extraction   Instruct   Mistral   Model-index   Mteb   Pytorch   Region:us   Safetensors   Sentence-transformers   Sharded   Tensorflow   Text-embeddings-inference
Model Card on HF 🤗: https://huggingface.co/intfloat/e5-mistral-7b-instruct 

E5 Mistral 7B Instruct Parameters and Internals

Model Type 
Multilingual
Additional Notes 
Fine-tuned on multilingual datasets.
Supported Languages 
primary (English), notes (Some multilingual capability.)
Input Output 
Input Format:
Queries with instructions
Output Format:
Embeddings
Performance Tips:
Use instructions to avoid performance degradation.
LLM NameE5 Mistral 7B Instruct
Repository 🤗https://huggingface.co/intfloat/e5-mistral-7b-instruct 
Model Size7b
Required VRAM14.2 GB
Updated2026-07-20
Maintainerintfloat
Model Typemistral
Instruction-BasedYes
Model Files  9.9 GB: 1-of-2   4.3 GB: 2-of-2   9.9 GB: 1-of-2   4.3 GB: 2-of-2
Supported Languagesen
Model ArchitectureMistralModel
Licensemit
Context Length32768
Model Max Length32768
Transformers Version4.34.0
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typefloat16