What are the hardware requirements for Chinese Text Correction 1.5B?

Chinese Text Correction 1.5B requires approximately 3.1 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Chinese Text Correction 1.5B and how large is it?

Chinese Text Correction 1.5B is developed by shibing624, a model with 1.5b 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 Chinese Text Correction 1.5B?

Chinese Text Correction 1.5B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Chinese Text Correction 1.5B by shibing624 — VRAM 3.1GB, 32K context

Name: Chinese Text Correction 1.5B
Author: shibing624

Chinese Text Correction 1.5B is an open-source language model by shibing624. Features: 1.5b LLM, VRAM: 3.1GB, Context: 32K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.15.

Base model:finetune:qwen/qwen2... Base model:qwen/qwen2.5-1.5b-i... Conversational Dataset:shibing624/chinese tex... Deploy:azure Endpoints compatible Instruct Qwen2 Region:us Safetensors Zh

Model Card on HF 🤗: https://huggingface.co/shibing624/chinese-text-correction-1.5b

Chinese Text Correction 1.5B Parameters and Internals

Model Type

text correction

Use Cases

Areas:

spelling correction, grammar correction

Primary Use Cases:

Chinese Text Correction (CTC), Chinese Spelling Correction (CSC)

Additional Notes

The model is part of the pycorrector project and can be used with or without it.

Supported Languages

zh (proficient)

Training Details

Data Sources:

shibing624/chinese_text_correction

Training Time:

9 days 8 hours

Hardware Used:

Tesla V100, 32 GB VRAM

Input Output

Input Format:

text input through encoder

Accepted Modalities:

text

Output Format:

corrected text output

LLM Name	Chinese Text Correction 1.5B
Repository 🤗	https://huggingface.co/shibing624/chinese-text-correction-1.5b
Base Model(s)	Qwen/Qwen2.5-1.5B-Instruct Qwen/Qwen2.5-1.5B-Instruct
Model Size	1.5b
Required VRAM	3.1 GB
Updated	2026-08-08
Maintainer	shibing624
Model Type	qwen2
Instruction-Based	Yes
Model Files	3.1 GB
Supported Languages	zh
Model Architecture	Qwen2ForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.37.2
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the Chinese Text Correction 1.5B 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 Chinese Text Correction 1.5B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen2.5 1.5B 1.3M Stretched	1280K / 3.1 GB	27	0
Merge Cord V1	128K / 3.1 GB	155	1
Gte Qwen2 1.5B Instruct	128K / 7.1 GB	346204	239
AceInstruct 1.5B	128K / 3.5 GB	139	22
... Abliterated TIES Qwen2.5 1.5B	128K / 3.5 GB	15	0
Saba1 1.8B	128K / 3.6 GB	6	1
Miniclaus Qw1.5B UNAMGS	128K / 3.5 GB	12	8
Bellatrix 1.5B XElite	128K / 3.5 GB	19	3
GCIRS Reasoning 1.5B R1	128K / 3.5 GB	9	1
...illed Scout 1.5B Instruct Gen1	128K / 3.5 GB	10	0

Note: green Score (e.g. "73.2") means that the model is better than shibing624/chinese-text-correction-1.5b.