What are the hardware requirements for SmolLM2 1.7B?

SmolLM2 1.7B requires approximately 3.4 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SmolLM2 1.7B and how large is it?

SmolLM2 1.7B is developed by HuggingFaceTB, a model with 1.7b 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 SmolLM2 1.7B?

SmolLM2 1.7B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

SmolLM2 1.7B by HuggingFaceTB — VRAM 3.4GB, 8K context

Name: SmolLM2 1.7B
Author: HuggingFaceTB

SmolLM2 1.7B is an open-source language model by HuggingFaceTB. Features: 1.7b LLM, VRAM: 3.4GB, Context: 8K, License: apache-2.0, LLM Explorer Score: 0.24.

Arxiv:2502.02737 Deploy:azure En Endpoints compatible Eval-results Llama Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B

SmolLM2 1.7B Parameters and Internals

Model Type

language model

Use Cases

Areas:

text rewriting, summarization, function calling

Applications:

research, commercial applications

Limitations:

Primarily understands and generates content in English., Generated content may not be factually accurate, logically consistent, or free from biases., Should be used as assistive tools rather than definitive sources of information.

Considerations:

Users should verify important information and critically evaluate any generated content.

Additional Notes

The instruct version of the model is tuned to support tasks beyond standard language modeling.

Supported Languages

en (primary)

Training Details

Data Sources:

FineWeb-Edu, DCLM, The Stack, new mathematics dataset, coding dataset

Data Volume:

11 trillion tokens

Methodology:

Supervised fine-tuning (SFT), Direct Preference Optimization (DPO)

Hardware Used:

256 H100 GPUs

Model Architecture:

Transformer decoder

Input Output

Accepted Modalities:

text

Performance Tips:

None specified.

LLM Name	SmolLM2 1.7B
Repository 🤗	https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B
Model Size	1.7b
Required VRAM	3.4 GB
Updated	2026-07-19
Maintainer	HuggingFaceTB
Model Type	llama
Model Files	3.4 GB
Supported Languages	en
Context Length	8k
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	8192
Model Max Length	8192
Transformers Version	4.40.1
Tokenizer Class	GPT2Tokenizer
Vocabulary Size	49152
Torch Data Type	bfloat16

Rank the SmolLM2 1.7B 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 SmolLM2 1.7B

Model	Likes	Downloads	VRAM
SmolTulu 1.7B Instruct	13	126	3 GB
SmolLM2 1.7B Bnb 4bit	3	165	1 GB

Best Alternatives to SmolLM2 1.7B

Best Alternatives	Context / RAM	Downloads	Likes
...FT Tulu3 Decontaminated Masked	16K / 3.4 GB	342	0
...B 16K SFT Tulu3 Decontaminated	16K / 3.4 GB	146	0
SmolLM2 1.7B 16K	16K / 3.4 GB	12	0
SmolLM2 1.7B Instruct 16K	16K / 3.4 GB	103	10
SmolLM2 1.7B Instruct	8K / 3.4 GB	154013	742
Yoru	8K / 3.4 GB	309	0
SmolLM2 1.7B Instruct	8K / 3.4 GB	4909	6
Augmented B933f090bb558b88	8K / 3.4 GB	574	0
...FT Tulu3 Decontaminated Masked	8K / 3.4 GB	274	0
Banglish Companion	8K / 6.8 GB	1254	0

Note: green Score (e.g. "73.2") means that the model is better than HuggingFaceTB/SmolLM2-1.7B.