What are the hardware requirements for TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2?

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 requires approximately 0.7 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 TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 and how large is it?

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 is developed by LoneStriker, a model with 503b 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 TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2?

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 by LoneStriker — VRAM 0.7GB, 2K context

Name: TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2
Author: LoneStriker

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 is an open-source language model by LoneStriker. Features: 503b LLM, VRAM: 0.7GB, Context: 2K, License: apache-2.0, Quantized, LLM Explorer Score: 0.09.

Dataset:bigcode/starcoderdata Dataset:cerebras/slimpajama-62... Dataset:openassistant/oasst to... En Endpoints compatible Exl2 Llama Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/LoneStriker/TinyLlama-1.1B-Chat-v0.3-4.0bpw-h6-exl2

TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 Parameters and Internals

Model Type

Chat, Text Generation

Additional Notes

TinyLlama can be easily integrated with projects based on Llama due to architectural and tokenizer similarities. It is optimized for efficiency with a compact size of 1.1 billion parameters.

Supported Languages

en (Proficient)

Training Details

Data Sources:

cerebras/SlimPajama-627B, bigcode/starcoderdata, OpenAssistant/oasst_top1_2023-08-25

Data Volume:

3 trillion tokens

Methodology:

Pretraining and finetuning

Training Time:

90 days estimated

Hardware Used:

16 A100-40G GPUs

Model Architecture:

Same as Llama 2

Input Output

Input Format:

chatml

Accepted Modalities:

text

Output Format:

text

LLM Name	TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2
Repository 🤗	https://huggingface.co/LoneStriker/TinyLlama-1.1B-Chat-v0.3-4.0bpw-h6-exl2
Model Size	503b
Required VRAM	0.7 GB
Updated	2026-08-04
Maintainer	LoneStriker
Model Type	llama
Model Files	0.7 GB
Supported Languages	en
Quantization Type	exl2
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.34.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	[PAD]
Vocabulary Size	32003
Torch Data Type	float32

Rank the TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2 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 TinyLlama 1.1B Chat V0.3 4.0bpw H6 EXL2

Best Alternatives	Context / RAM	Downloads	Likes
... 1.1B Chat V0.3 8.0bpw H8 EXL2	2K / 1.2 GB	13	1
... 1.1B Chat V0.3 5.0bpw H6 EXL2	2K / 0.8 GB	5	0
TinyLlama 1.1bee	2K / 4.4 GB	18	1
TinyLlama 1.1B Chat V0.1	2K / 4.4 GB	603	57
TinyLlama 1.1B Chat V0.3	2K / 4.4 GB	1251	44
TinyLlama 1.1B Chat V0.2	2K / 4.4 GB	287	17
...1B Intermediate Step 240K 503B	2K / 4.4 GB	171	21
TinyLlama 1.1B Orca Gpt4	2K / 2.2 GB	7	1
TinyLlama 1.1B Dolly	2K / 2.2 GB	20	0
TinyLlama 1.1B	2K / 2.2 GB	11	2

Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/TinyLlama-1.1B-Chat-v0.3-4.0bpw-h6-exl2.