What are the hardware requirements for Lab21 LoRA CSKH?

Lab21 LoRA CSKH requires approximately 0.1 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Lab21 LoRA CSKH and how large is it?

Lab21 LoRA CSKH is developed by QunAnh, a model with 4b 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 Lab21 LoRA CSKH?

Lab21 LoRA CSKH is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Lab21 LoRA CSKH by QunAnh — VRAM 0.1GB

Name: Lab21 LoRA CSKH
Author: QunAnh

Lab21 LoRA CSKH is an open-source language model by QunAnh. Features: 4b LLM, VRAM: 0.1GB, LLM Explorer Score: 0.28.

Adapter Base model:adapter:unsloth/qwe... Base model:unsloth/qwen3.5-4b Conversational Finetuned Lora Peft Region:us Safetensors Ticket-triage Vi

Model Card on HF 🤗: https://huggingface.co/QunAnh/Lab21-LoRA-CSKH

Lab21 LoRA CSKH Parameters and Internals

LLM Name	Lab21 LoRA CSKH
Repository 🤗	https://huggingface.co/QunAnh/Lab21-LoRA-CSKH
Base Model(s)	unsloth/Qwen3.5-4B unsloth/Qwen3.5-4B
Model Size	4b
Required VRAM	0.1 GB
Updated	2026-10-11
Maintainer	QunAnh
Model Files	0.1 GB
Supported Languages	vi
Model Architecture	Adapter
Model Max Length	262144
Is Biased	none
Tokenizer Class	Qwen3_5Tokenizer
Padding Token	<\|vision_pad\|>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	out_proj\|gate_proj\|v_proj\|down_proj\|up_proj\|in_proj_a\|k_proj\|in_proj_qkv\|in_proj_z\|o_proj\|q_proj\|in_proj_b
LoRA Alpha	32
LoRA Dropout	0
R Param	16
Errors	replace

Rank the Lab21 LoRA CSKH 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 Lab21 LoRA CSKH

Best Alternatives	Context / RAM	Downloads	Likes
Kev 4B	0K / 0.1 GB	16100	101
Qwen Las V0.1	0K / 0 GB	6	0
Visual Jev 4B Answer Sft	0K / 0.1 GB	466	2
Medjev 4B Lora	0K / 0.1 GB	41	2
Imajev 4B	0K / 0.5 GB	64	3
...c Qwen3.5 4B Lora Experimental	0K / 0.2 GB	24	1
...ate Field Intelligence Adapter	0K / 0.1 GB	44	1
...wen3.5 4B Turkish Concise Lora	0K / 0.1 GB	12	1
...oolTrajectories InjectionAware	0K / 0.7 GB	14	1
...Chess Reasoner Qwen3.5 4B Lora	0K / 0.1 GB	14	0

Note: green Score (e.g. "73.2") means that the model is better than QunAnh/Lab21-LoRA-CSKH.