What are the hardware requirements for TinyLlama 1.1B Step 50K 105B?

TinyLlama 1.1B Step 50K 105B requires approximately 4.4 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 Step 50K 105B and how large is it?

TinyLlama 1.1B Step 50K 105B is developed by TinyLlama, a model with 105b 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 Step 50K 105B?

TinyLlama 1.1B Step 50K 105B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

TinyLlama 1.1B Step 50K 105B by TinyLlama — VRAM 4.4GB, 2K context

Name: TinyLlama 1.1B Step 50K 105B
Author: TinyLlama

TinyLlama 1.1B Step 50K 105B is an open-source language model by TinyLlama. Features: 105b LLM, VRAM: 4.4GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.13.

Dataset:bigcode/starcoderdata Dataset:cerebras/slimpajama-62... En Endpoints compatible Llama Pytorch Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TinyLlama/TinyLlama-1.1B-step-50K-105b

TinyLlama 1.1B Step 50K 105B Parameters and Internals

Model Type

text generation

Use Cases

Areas:

many open-source projects, applications demanding restricted computation and memory footprint

Supported Languages

en (Low)

Training Details

Data Sources:

cerebras/SlimPajama-627B, bigcode/starcoderdata

Data Volume:

3 trillion tokens

Training Time:

90 days

Hardware Used:

16 A100-40G GPUs

Model Architecture:

Same architecture and tokenizer as Llama 2

Input Output

Accepted Modalities:

text

Output Format:

text

Performance Tips:

Use torch_dtype=torch.float16 and device_map='auto'.

Release Notes

Version:

intermediate 50K steps, 105B tokens

Date:

2023-09-04

Notes:

Initial intermediate checkpoint release.

LLM Name	TinyLlama 1.1B Step 50K 105B
Repository 🤗	https://huggingface.co/TinyLlama/TinyLlama-1.1B-step-50K-105b
Model Size	105b
Required VRAM	4.4 GB
Updated	2026-07-24
Maintainer	TinyLlama
Model Type	llama
Model Files	4.4 GB 4.4 GB
Supported Languages	en
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.31.0.dev0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float32

Rank the TinyLlama 1.1B Step 50K 105B 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

Quantized Models of the TinyLlama 1.1B Step 50K 105B

Model	Likes	Downloads	VRAM
...ma 1.1B Step 50K 105B AWQ 4bit	0	6	0 GB

Best Alternatives to TinyLlama 1.1B Step 50K 105B

Best Alternatives	Context / RAM	Downloads	Likes
Anubis Pro 105B V1	128K / 210.7 GB	4	32
TinyLlama	2K / 2.2 GB	5	0
...yLlama 1.1B Step 50K 105B ONNX	2K / GB	10	1
Tiny Llama 1.1b Odia Ext	2K / 2.2 GB	6	0
...ma 1.1B Step 50K 105B AWQ 4bit	2K / 0.8 GB	6	0
...a 1.1B Step 50K 105B Gptq 4bit	2K / 0.8 GB	14	1

Note: green Score (e.g. "73.2") means that the model is better than TinyLlama/TinyLlama-1.1B-step-50K-105b.