What are the hardware requirements for LiteLlama 460M 1T?

LiteLlama 460M 1T requires approximately 0.9 GB of VRAM and supports a context window of 1K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed LiteLlama 460M 1T and how large is it?

LiteLlama 460M 1T is developed by ahxt, a model with 460m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does LiteLlama 460M 1T perform on standard benchmarks?

LiteLlama 460M 1T has the following published scores: MMLU 25.96. Compare against reference models on this page or on the LLM Explorer leaderboards.

LiteLlama 460M 1T by ahxt — VRAM 0.9GB, 1K context

Name: LiteLlama 460M 1T
Author: ahxt

LiteLlama 460M 1T is an open-source language model by ahxt. Features: 460m LLM, VRAM: 0.9GB, Context: 1K, License: mit, LLM Explorer Score: 0.16, Arc: 24.8, HellaSwag: 38.4, MMLU: 26.

Dataset:redpajama Deploy:azure Deploy:sagemaker En Endpoints compatible Litellama Llama Llama2 Llama2 architecture Pytorch Region:us

Model Card on HF 🤗: https://huggingface.co/ahxt/LiteLlama-460M-1T

LiteLlama 460M 1T Parameters and Internals

Model Type

text generation

Use Cases

Areas:

Research, Commercial Applications

Applications:

Text Generation

Additional Notes

The model is compact and experimental with significantly reduced parameter size compared to its base model LLaMa 2.

Training Details

Data Sources:

RedPajama

Data Volume:

0.98 trillion tokens

Context Length:

1024

LLM Name	LiteLlama 460M 1T
Repository 🤗	https://huggingface.co/ahxt/LiteLlama-460M-1T
Model Size	460m
Required VRAM	0.9 GB
Updated	2026-07-23
Maintainer	ahxt
Model Type	llama
Model Files	0.9 GB
Supported Languages	en
Model Architecture	LlamaForCausalLM
License	mit
Context Length	1024
Model Max Length	1024
Transformers Version	4.31.0
Tokenizer Class	GPT2Tokenizer
Beginning of Sentence Token	<\|endoftext\|>
End of Sentence Token	<\|endoftext\|>
Unk Token	<\|endoftext\|>
Vocabulary Size	50304
Torch Data Type	bfloat16
Errors	replace

Rank the LiteLlama 460M 1T 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 LiteLlama 460M 1T

Best Alternatives	Context / RAM	Downloads	Likes
LLAMA 2 460M	4K / 1.8 GB	5	0
TeenyTinyLlama Doc 460M Chat	2K / 1.9 GB	10	0
Doctor Llama Chat	2K / 1.9 GB	7	0
Doctor Llama 460M	2K / 0 GB	9	0
TeenyTinyLlama Doc 460M	2K / 0 GB	8	0
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
DiscordLLama 460M	2K / 0.8 GB	9	0
Lite Vortex	1K / 0.9 GB	17	2