What are the hardware requirements for SmolLM2 1.7B Instruct?

SmolLM2 1.7B Instruct 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 Instruct and how large is it?

SmolLM2 1.7B Instruct 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.

How does SmolLM2 1.7B Instruct perform on standard benchmarks?

SmolLM2 1.7B Instruct has the following published scores: LMSYS ELO 1062. Compare against reference models on this page or on the LLM Explorer leaderboards.

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

Name: SmolLM2 1.7B Instruct
Author: HuggingFaceTB

SmolLM2 1.7B Instruct is an open-source language model by HuggingFaceTB. Features: 1.7b LLM, VRAM: 3.4GB, Context: 8K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.4, ELO: 1062.

Arxiv:2502.02737 Base model:huggingfacetb/smoll... Base model:quantized:huggingfa... Conversational En Endpoints compatible Instruct Llama Onnx Region:us Safetensors Tensorboard Transformers.js

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

SmolLM2 1.7B Instruct Benchmarks

LMSys ELO 1062 vs 1272 (so35) -16.5%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

SmolLM2 1.7B Instruct Parameters and Internals

Model Type

text generation, instruction following

Use Cases

Areas:

text generation, instruction following, text rewriting, summarization, function calling

Applications:

educational tools, coding assistance, customer support, chatbots, language learning

Primary Use Cases:

text and instruction generation

Limitations:

primarily understands and generates content in English, generated content may not always be factually accurate or free from bias

Considerations:

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

Supported Languages

en (main)

Training Details

Data Sources:

FineWeb-Edu, DCLM, The Stack, new mathematics and coding datasets

Data Volume:

11T tokens

Methodology:

Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO)

Hardware Used:

256 H100 GPUs

Model Architecture:

Transformer decoder

Input Output

Input Format:

Expected input format includes system and user prompts

Accepted Modalities:

text

Output Format:

text

Performance Tips:

Ensure queries are clear and check model's language support for optimal results.

LLM Name	SmolLM2 1.7B Instruct
Repository 🤗	https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct
Base Model(s)	HuggingFaceTB/SmolLM2-1.7B HuggingFaceTB/SmolLM2-1.7B
Model Size	1.7b
Required VRAM	3.4 GB
Updated	2026-07-14
Maintainer	HuggingFaceTB
Model Type	llama
Instruction-Based	Yes
Model Files	3.4 GB 0.0 GB
Supported Languages	en
Context Length	8k
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	8192
Model Max Length	8192
Transformers Version	4.42.3
Tokenizer Class	GPT2Tokenizer
Padding Token	<\|im_end\|>
Vocabulary Size	49152
Torch Data Type	bfloat16

Rank the SmolLM2 1.7B Instruct 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 Instruct

Model	Likes	Downloads	VRAM
TwIL LM	10	519	1 GB
SmolLM2 1.7B Instruct Q8 Mlx	5	258	1 GB
SmolLM2 1.7B Instruct Bnb 4bit	2	407	1 GB

Best Alternatives to SmolLM2 1.7B Instruct

Best Alternatives	Context / RAM	Downloads	Likes
SmolLM2 1.7B Instruct 16K	16K / 3.4 GB	103	10
Yoru	8K / 3.4 GB	309	0
SmolLM2 1.7B Instruct	8K / 3.4 GB	4909	6
...ygnis Alpha 1.7B V0.1 Instruct	8K / 3.4 GB	11	2
SmolTulu 1.7B Reinforced	8K / 3.4 GB	12	5
SmolLM2 1.7 Persona	8K / 3.5 GB	7	0
Superthoughts Lite V1	8K / 3.4 GB	92	3
Penny 1.7B	8K / 6.9 GB	18	24
...ghts Lite 1.8B Experimental O1	8K / 3.6 GB	11	2
Cllm 1.0.0 340000 Instruct	8K / 3.5 GB	5	0