Breeze-7B-Instruct-v1_0 Open Source Language Model - Free Deployment, Supports Traditional Chinese Q&A and Multi-round Conversations

Breeze 7B Instruct V1 0

Developed by MediaTek-Research

Breeze-7B-Instruct is a Traditional Chinese optimized language model based on Mistral-7B, specifically designed for instruction-following tasks, supporting scenarios such as Q&A and multi-turn dialogues.

Large Language Model

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Traditional Chinese optimization #Extended vocabulary acceleration #Multi-turn dialogue support

Downloads 1,388

Release Time : 3/5/2024

Model Overview

This model is the instruction-tuned version in MediaTek's Breeze-7B series, optimized for Traditional Chinese applications, and can be directly used for common tasks like Q&A, retrieval-augmented generation, multi-turn dialogues, and text summarization.

Model Features

Extended vocabulary optimization

Vocabulary expanded from 32k to 62k, significantly improving Traditional Chinese processing efficiency with inference speeds twice as fast as similar models.

Long context support

Supports context lengths of up to 8k tokens, suitable for processing long documents and complex dialogue scenarios.

Multi-turn dialogue capability

Specifically optimized for multi-turn dialogue interactions, supporting coherent contextual understanding.

Model Capabilities

Q&A systems

Multi-turn dialogues

Text summarization

Retrieval-augmented generation

Traditional Chinese text generation

English text generation

Use Cases

Intelligent customer service

Traditional Chinese customer support

Handling customer service inquiries from users in Taiwan

Achieved a score of 6.0 in the MT-Bench-tw dialogue evaluation

Educational assistance

Traditional Chinese learning assistant

Helping students answer questions related to Traditional Chinese

Achieved 42.67% accuracy in the TMMLU+ knowledge evaluation

🚀 Model Card for MediaTek Research Breeze-7B-Instruct-v1_0

MediaTek Research Breeze-7B is a language model family based on Mistral-7B, specifically designed for Traditional Chinese use. It offers two main models: Breeze-7B-Base and Breeze-7B-Instruct, with the current release version being v1.0, which shows significant performance improvements in both English and Traditional Chinese.

🚀 Quick Start

For details of this model, please read our paper. You can also try the demo here.

✨ Features

Breeze-7B-Base-v1_0

Expanded the vocabulary dictionary size from 32k to 62k to better support Traditional Chinese.
8k-token context length.

Breeze-7B-Instruct-v1_0

Expanded the vocabulary dictionary size from 32k to 62k to better support Traditional Chinese.
8k-token context length.
Supports multi-turn dialogue (without special handling for harmfulness).

📚 Documentation

Model Details

Breeze-7B-Base-v1_0

Finetuned from: mistralai/Mistral-7B-v0.1
Model type: Causal decoder-only transformer language model
Language: English and Traditional Chinese (zh-tw)

Breeze-7B-Instruct-v1_0

Finetuned from: MediaTek-Research/Breeze-7B-Base-v1_0
Model type: Causal decoder-only transformer language model
Language: English and Traditional Chinese (zh-tw)

Base Model Performance

We compared Breeze-7B-Base-v1_0 with other open-source base language models of similar parameter size. The evaluation was conducted using the code revised from EleutherAI/lm-evaluation-harness.

Models	#Parameters	↑ TMMLU+ (ACC)	DRCD (EM)	Table (ACC)	MMLU (ACC)
		TC, Knowledge	TC, Reasoning	TC, Reasoning	EN, Knowledge
		5 shot	3 shot	5 shot	5 shot
Yi-6B	6B	49.63	76.61	34.72	65.35
Qwen1.5-7B	7B	46.59	74.41	30.56	63.07
Breeze-7B-Base-v1_0	7B	42.67	80.61	31.99	61.24
Mistral-7B-v0.1	7B	36.93	79.27	27.78	64.89

Instruction-tuned Model Performance

We compared Breeze-7B-Instruct-v1_0 with other open-source instruction-tuned language models of similar parameter size. Different evaluation codes were used for different benchmarks.

Models	#Parameters	↑ MT-Bench-tw (Score)	TMMLU+ (ACC)	Table (ACC)	MT-Bench (Score)	MMLU (ACC)
		TC, Chat	TC, Knowledge	TC, Reasoning	EN, Chat	EN, Knowledge
		0 shot	0 shot	0 shot	0 shot	0 shot
GPT-3.5-Turbo		7.1	43.56	45.14	7.9	67.09
Qwen1.5-7B-Chat	7B	6.4	45.65	34.72	7.6	61.85
Breeze-7B-Instruct-v1_0	7B	6.0	42.67	39.58	7.4	61.73
Mistral-7B-v0.2-Instruct	7B	5.6	34.95	33.33	7.6	59.97
Yi-6B-Chat	6B	5.0	44.79	25.69	6.0	59.45
Taiwan-LLM-13B-v2.0-chat	13B	5.0	29.47	23.61	N/A*	50.50
Taiwan-LLM-7B-v2.1-chat	7B	4.2	28.08	31.25	N/A*	42.72

* Taiwan-LLM models respond to multi-turn questions (English) in Traditional Chinese.

Details on MT-Bench-tw (0 shot): Models	STEM	Extraction	Reasoning	Math	Coding	Roleplay	Writing	Humanities	AVG
GPT-3.5-Turbo	7.8	6.1	5.1	6.4	6.2	8.7	7.4	9.3	7.1
Qwen1.5-7B-Chat	9	5.6	4.7	2.8	3.7	8.0	8.0	9.4	6.4
Breeze-7B-Instruct-v1_0	7.8	5.2	4.2	4.2	4.1	7.6	5.9	9.1	6.0
Mistral-7B-v0.2-Instruct	6.9	4.6	4.3	3.3	4.4	7.2	6.2	7.8	5.6
Yi-6B-Chat	7.3	2.7	3.1	3.3	2.3	7.2	5.2	8.8	5.0
Taiwan-LLM-13B-v2.0-chat	6.1	3.4	4.1	2.3	3.1	7.4	6.6	6.8	5.0
Taiwan-LLM-7B-v2.1-chat	5.2	2.6	2.3	1.2	3.4	6.6	5.7	6.8	4.2

Details on TMMLU+ (0 shot): Model	STEM	Social Science	Humanities	Other	AVG
GPT-3.5-Turbo	41.58	48.52	40.96	43.18	43.56
Qwen1.5-7B-Chat	41.48	51.66	44.05	45.40	45.65
Breeze-7B-Instruct-v1_0	36.46	48.38	45.11	40.75	42.67
Mistral-7B-v0.2-Instruct	32.79	38.05	34.89	34.04	34.94
Yi-6B-Chat	37.80	51.74	45.36	44.25	44.79
Taiwan-LLM-13B-v2.0-chat	27.74	33.69	27.03	29.43	29.47
Taiwan-LLM-7B-v2.1-chat	25.58	31.76	27.36	27.61	28.08

Inference Performance

In this test, we used the first 700 characters of the web article as the input and asked the model to rewrite the same article. All inferences were run on 2 RTX A6000 GPUs.

Models	↓ Inference Time (sec)	Estimated Max Input Length (Char)
Qwen1.5-7B-Chat	9.35	38.9k
Yi-6B-Chat	10.62	5.2k
Breeze-7B-Instruct-v1_0	10.74	11.1k
Mistral-7B-Instruct-v0.2	20.48	5.1k
Taiwan-LLM-7B-v2.1-chat	26.26	2.2k

📄 License

This model is licensed under the Apache-2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご