What are the hardware requirements for MiniLLM 0.2B WithWudao?

MiniLLM 0.2B WithWudao requires approximately 0.9 GB of VRAM and supports a context window of 1K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed MiniLLM 0.2B WithWudao and how large is it?

MiniLLM 0.2B WithWudao is developed by Tongjilibo, a model with 0.2b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate MiniLLM 0.2B WithWudao?

MiniLLM 0.2B WithWudao is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MiniLLM 0.2B WithWudao by Tongjilibo — VRAM 0.9GB, 1K context

Name: MiniLLM 0.2B WithWudao
Author: Tongjilibo

MiniLLM 0.2B WithWudao is an open-source language model by Tongjilibo. Features: 0.2b LLM, VRAM: 0.9GB, Context: 1K, License: apache-2.0, LLM Explorer Score: 0.11.

Autotrain compatible Endpoints compatible Llama Pytorch Region:us

Model Card on HF 🤗: https://huggingface.co/Tongjilibo/MiniLLM-0.2B-WithWudao

MiniLLM 0.2B WithWudao Parameters and Internals

Model Type

Text Generation, Chatbot

Use Cases

Areas:

Research, Shot-term chatbot functionalities

Primary Use Cases:

Simple chat or conversation scenarios

Limitations:

Limited to simple chat tasks, Cannot handle complex queries

Considerations:

The model's chat capabilities are limited due to size constraints and available training data.

Additional Notes

The model is aimed at demonstrating feasible LLM deployment with budget-friendly resources

Supported Languages

zh (中文)

Training Details

Data Sources:

Wiki中文百科, BaiduBaiKe, C4_zh, WuDaoCorpora, shibing624/medical

Data Volume:

634 Billion Tokens

Methodology:

Pre-training followed by instruction fine-tuning

Training Time:

NoWudao: 20h, WithWudao: 3.79d

Hardware Used:

4×A800 (80G)

Model Architecture:

bert4torch framework used for training

Input Output

Input Format:

Pre-tokenized input suitable for Llama models

Accepted Modalities:

Text

Output Format:

Textual responses

Performance Tips:

Recommended to follow the installation and usage guidelines for optimal performance.

Release Notes

Version:

20240316

Notes:

Initial model release including pre-trained versions with and without Wudao datasets.

LLM Name	MiniLLM 0.2B WithWudao
Repository 🤗	https://huggingface.co/Tongjilibo/MiniLLM-0.2B-WithWudao
Model Size	0.2b
Required VRAM	0.9 GB
Updated	2025-03-22
Maintainer	Tongjilibo
Model Type	llama
Model Files	0.9 GB 0.9 GB
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	1024
Model Max Length	1024
Tokenizer Class	ChatGLMTokenizer
Vocabulary Size	64793

Rank the MiniLLM 0.2B WithWudao 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 MiniLLM 0.2B WithWudao

Best Alternatives	Context / RAM	Downloads	Likes
Clef 0.2B	4K / 0.4 GB	122	2
Informer 0.2B 4K	4K / 0.5 GB	9	4
MiniLLM 0.2B SFT	1K / 0.9 GB	9	1
MiniLLM 0.2B WithWudao SFT	1K / 0.9 GB	8	1
MiniLLM 0.2B NoWudao	1K / 0.9 GB	31	1
MiniLLM 0.2B Base	1K / 0.9 GB	11	2
MiniLLM 0.2B NoWudao Base	1K / 0.9 GB	11	1

Note: green Score (e.g. "73.2") means that the model is better than Tongjilibo/MiniLLM-0.2B-WithWudao.