What are the hardware requirements for TinyStoriesChinese 110M?

TinyStoriesChinese 110M requires approximately 0.4 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 TinyStoriesChinese 110M and how large is it?

TinyStoriesChinese 110M is developed by fzmnm, a model with 110m 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 TinyStoriesChinese 110M?

TinyStoriesChinese 110M is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

TinyStoriesChinese 110M by fzmnm — VRAM 0.4GB, 1K context

Name: TinyStoriesChinese 110M
Author: fzmnm

TinyStoriesChinese 110M is an open-source language model by fzmnm. Features: 110m LLM, VRAM: 0.4GB, Context: 1K, License: cc, LLM Explorer Score: 0.12.

Arxiv:2203.15556 Dataset:adam89/tinystorieschin... Endpoints compatible Llama Pytorch Region:us Zh

Model Card on HF 🤗: https://huggingface.co/fzmnm/TinyStoriesChinese-110M

TinyStoriesChinese 110M Parameters and Internals

Model Type

text generation

Use Cases

Areas:

Educational purposes, Machine learning training for beginners

Applications:

Generation of short Chinese stories

Primary Use Cases:

Text generation for educational tools

Limitations:

Consistency and depth of logical capabilities vary

Considerations:

Focused on simple, narrowly defined tasks; ideal for experimentation

Additional Notes

Demonstrates potential of smaller models to produce coherent text without extensive computational resources.

Supported Languages

zh (Proficient)

Training Details

Data Sources:

adam89/TinyStoriesChinese

Data Volume:

2G tokens

Methodology:

Standard llama2 format Transformer with 12 layers, 12 attention heads, and a hidden size of 768. The model utilizes a 1024 context window with Relative Positional Encodings (RoPE) and a vocabulary size of 5,000.

Context Length:

1024

Training Time:

87 hours

Hardware Used:

NVIDIA RTX 2080 Super with 8 GB RAM

Model Architecture:

Standard llama2 format Transformer

Input Output

Input Format:

Text prompt

Accepted Modalities:

text

Output Format:

Generated story text

LLM Name	TinyStoriesChinese 110M
Repository 🤗	https://huggingface.co/fzmnm/TinyStoriesChinese-110M
Model Size	110m
Required VRAM	0.4 GB
Updated	2026-07-22
Maintainer	fzmnm
Model Type	llama
Model Files	0.4 GB
Supported Languages	zh
Model Architecture	LlamaForCausalLM
License	cc
Context Length	1024
Model Max Length	1024
Transformers Version	4.41.2
Tokenizer Class	LlamaTokenizer
Vocabulary Size	5000

Rank the TinyStoriesChinese 110M 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 TinyStoriesChinese 110M

Best Alternatives	Context / RAM	Downloads	Likes
TinyBalls 110M V1	32K / 0.5 GB	314	1
BigWeave V20 110B	4K / 220 GB	51	0
...10M Experimental V0.1 Instruct	1K / 0.2 GB	36	1
HelpingAI 110M	1K / 0.2 GB	508	2
Tinybooks 110M Idiot Iter1000	1K / 0.4 GB	6	0
Tinyllama 110M Fp32	1K / 0.4 GB	1480	0
Llama 110M	1K / 0.2 GB	11	0
Tinyllama 110M	1K / 0.2 GB	866	6

Note: green Score (e.g. "73.2") means that the model is better than fzmnm/TinyStoriesChinese-110M.