What are the hardware requirements for TeenyTinyLlama 460M AWQ?

TeenyTinyLlama 460M AWQ requires approximately 0.3 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 TeenyTinyLlama 460M AWQ and how large is it?

TeenyTinyLlama 460M AWQ is developed by nicholasKluge, a model with 460m 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 TeenyTinyLlama 460M AWQ?

TeenyTinyLlama 460M AWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

TeenyTinyLlama 460M AWQ by nicholasKluge — VRAM 0.3GB, 2K context

Name: TeenyTinyLlama 460M AWQ
Author: nicholasKluge

TeenyTinyLlama 460M AWQ is an open-source language model by nicholasKluge. Features: 460m LLM, VRAM: 0.3GB, Context: 2K, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.11.

Arxiv:2401.16640 4-bit Awq Base model:nicholaskluge/teeny... Base model:quantized:nicholask... Co2 eq emissions Dataset:nicholaskluge/pt-corpu... Endpoints compatible Instruct Llama Pt Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/nicholasKluge/TeenyTinyLlama-460m-awq

TeenyTinyLlama 460M AWQ Parameters and Internals

Model Type

Transformer-based, Text-generation

Use Cases

Areas:

Research

Primary Use Cases:

Research challenges related to developing language models for low-resource languages

Limitations:

Not intended for deployment, Not suitable for translation or generating text in other languages

Additional Notes

Quantized version using AutoAWQ, making it 80% lighter, 20% faster with minimal performance loss.

Supported Languages

Portuguese (only)

Training Details

Data Sources:

Pt-Corpus Instruct (6.2B tokens)

Data Volume:

6.2B tokens

Methodology:

Causal language modeling

Context Length:

2048

Training Time:

~ 280 hours

Hardware Used:

1 NVIDIA A100-SXM4-40GB

Model Architecture:

Transformer-based

Input Output

Accepted Modalities:

Text

Output Format:

Text generation

Performance Tips:

Using quantized models required the installation of `autoawq==0.1.7`. A GPU is required to run the AWQ-quantized models.

LLM Name	TeenyTinyLlama 460M AWQ
Repository 🤗	https://huggingface.co/nicholasKluge/TeenyTinyLlama-460m-awq
Base Model(s)	TeenyTinyLlama 460M nicholasKluge/TeenyTinyLlama-460m
Model Size	460m
Required VRAM	0.3 GB
Updated	2026-08-03
Maintainer	nicholasKluge
Model Type	llama
Instruction-Based	Yes
Model Files	0.3 GB
Supported Languages	pt
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Padding Token	<pad>
Vocabulary Size	32000
Torch Data Type	float16

Rank the TeenyTinyLlama 460M AWQ 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 TeenyTinyLlama 460M AWQ

Best Alternatives	Context / RAM	Downloads	Likes
TeenyTinyLlama 460M Chat AWQ	2K / 0.3 GB	4	1
TeenyTinyLlama 460M Chat	2K / 0 GB	90	3
TeenyTinyLlama 460M	2K / 1.9 GB	145	12
...60M Experimental Ptbr Instruct	2K / 0.9 GB	8	4

Note: green Score (e.g. "73.2") means that the model is better than nicholasKluge/TeenyTinyLlama-460m-awq.