What are the hardware requirements for CSTinyLlama 1.2B?

CSTinyLlama 1.2B requires approximately 4.9 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed CSTinyLlama 1.2B and how large is it?

CSTinyLlama 1.2B is developed by BUT-FIT, a model with 1.2b 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 CSTinyLlama 1.2B?

CSTinyLlama 1.2B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

CSTinyLlama 1.2B by BUT-FIT — VRAM 4.9GB, 2K context

Name: CSTinyLlama 1.2B
Author: BUT-FIT

CSTinyLlama 1.2B is an open-source language model by BUT-FIT. Features: 1.2b LLM, VRAM: 4.9GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.12.

Cs Dataset:but-fit/adult content ... Dataset:but-fit/but-lcc Endpoints compatible Llama Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/BUT-FIT/CSTinyLlama-1.2B

CSTinyLlama 1.2B Parameters and Internals

Model Type

language model, text generation

Additional Notes

This model is primarily for Czech language processing and includes a tokenizer built with a vocabulary swap method.

Supported Languages

cs (high)

Training Details

Data Sources:

BUT-FIT/BUT-LCC, BUT-FIT/adult_content_classifier_dataset

Data Volume:

168 billion tokens

Methodology:

Continuously pretrained using a vocabulary swap method

Context Length:

2048

Hardware Used:

Karolina cluster

Model Architecture:

Derived from TinyLlama-2.5T

Input Output

Accepted Modalities:

text

Output Format:

text

LLM Name	CSTinyLlama 1.2B
Repository 🤗	https://huggingface.co/BUT-FIT/CSTinyLlama-1.2B
Model Size	1.2b
Required VRAM	4.9 GB
Updated	2026-08-25
Maintainer	BUT-FIT
Model Type	llama
Model Files	4.9 GB
Supported Languages	cs
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.37.0.dev0
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	[EOS]
Vocabulary Size	64002
Torch Data Type	float32

Rank the CSTinyLlama 1.2B 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 CSTinyLlama 1.2B

Best Alternatives	Context / RAM	Downloads	Likes
Mazgu Llama V3 Merged	128K / 2.5 GB	301	0
Assn2 Sft Train5	128K / 2.5 GB	7	0
Zay Llama1b DPO Harmful PathD	128K / 2.5 GB	1	1
Akshar Nano V2	128K / 2.5 GB	15	0
Akshar Nano	128K / 2.5 GB	10	0
Second Tts Try	128K / 2.5 GB	39	0
...18 27e6 491c B3ab Ea03a5b93541	128K / 4.9 GB	10	0
...e9 9a4a 4ead 84f4 71761ee932ca	128K / 4.9 GB	6	0
SemAFacet SFT Merged 10K	128K / 4.9 GB	9	0
Llamainstructbadendings	128K / 2.5 GB	13	0

Note: green Score (e.g. "73.2") means that the model is better than BUT-FIT/CSTinyLlama-1.2B.