LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Breeze 7B Instruct V0.1 by MediaTek-Research

 »  All LLMs  »  MediaTek-Research  »  Breeze 7B Instruct V0.1   URL Share it on

  1190   
  Arxiv:2403.02712   Autotrain compatible   Conversational   En   Endpoints compatible   Instruct   Mistral   Pytorch   Region:us   Safetensors   Sharded   Tensorflow   Zh
Model Card on HF 🤗: https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0_1 

Breeze 7B Instruct V0.1 Benchmarks

LLME Score: 0.18574
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Breeze 7B Instruct V0.1 (MediaTek-Research/Breeze-7B-Instruct-v0_1)

Breeze 7B Instruct V0.1 Parameters and Internals

Model Type 
Causal decoder-only transformer language model
Use Cases 
Applications:
Q&A, RAG, multi-round chat, summarization
Considerations:
Breeze-7B-Instruct can be used as-is for commonly seen tasks without any special handling for harmfulness.
Additional Notes 
The model has expanded vocabulary for Traditional Chinese, aiming to improve inference speed.
Supported Languages 
zh-tw (Traditional Chinese), en (English)
Training Details 
Data Sources:
Mistral-7B-v0.1
Methodology:
Fine-tuning the base model Breeze-7B-Base.
Context Length:
8000
Model Architecture:
Causal decoder-only transformer
Input Output 
Input Format:
Expected prompts for multi-turn dialogue.
Accepted Modalities:
text
Output Format:
Textual output responses.
Performance Tips:
Incorporate `flash-attention2` for faster inference.
Release Notes 
Version:
v0.1
Notes:
Initial release of Breeze-7B-Instruct, supporting Traditional Chinese and English with impressive benchmark results.
LLM NameBreeze 7B Instruct V0.1
Repository 🤗https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0_1 
Model Size7b
Required VRAM15 GB
Updated2025-02-05
MaintainerMediaTek-Research
Model Typemistral
Instruction-BasedYes
Model Files  10.0 GB: 1-of-2   5.0 GB: 2-of-2   10.0 GB: 1-of-2   5.0 GB: 2-of-2
Supported Languageszh en
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.34.1
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size61952
Torch Data Typebfloat16

Quantized Models of the Breeze 7B Instruct V0.1

Model
Likes
Downloads
VRAM
Breeze 7B Instruct V0.1 GPTQ23764 GB

Best Alternatives to Breeze 7B Instruct V0.1

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB522710
SpydazWeb AI HumanAI RP512K / 14.4 GB51
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
... Summarize 64K QLoRANET Merged128K / 4.1 GB760
...1 Summarize 64K LoRANET Merged128K / 14.4 GB50
Mistral 7B Instruct V0.232K / 14.4 GB33160952630
Mistral 7B Instruct V0.132K / 14.4 GB12462781551
...ity Instruct 7M Gen Mistral 7B32K / 14.4 GB37145
...ty Instruct 3M 0613 Mistral 7B32K / 14.4 GB641511
Note: green Score (e.g. "73.2") means that the model is better than MediaTek-Research/Breeze-7B-Instruct-v0_1.

Rank the Breeze 7B Instruct 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? 42577 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.
Release v20241227
  Support LLM Explorer