LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

SmolLM 1.7B Instruct by HuggingFaceTB

 »  All LLMs  »  HuggingFaceTB  »  SmolLM 1.7B Instruct   URL Share it on

  42724   
  Alignment-handbook   Autotrain compatible Base model:huggingfacetb/smoll... Base model:quantized:huggingfa...   Conversational Dataset:bigcode/self-oss-instr... Dataset:huggingfacetb/everyday... Dataset:magpie-align/magpie-pr...   Dataset:teknium/openhermes-2.5   En   Endpoints compatible   Instruct   Llama   Onnx   Region:us   Safetensors   Sft   Trl
Model Card on HF 🤗: https://huggingface.co/HuggingFaceTB/SmolLM-1.7B-Instruct 

SmolLM 1.7B Instruct Benchmarks

MMLU Pro: 1.85
GPQA: 1.34
MUSR: 2.08
BBH: 2.08
IFEval: 23.48 vs 88 (so35)-73.3%
LLME Score: 0.33822
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SmolLM 1.7B Instruct (HuggingFaceTB/SmolLM-1.7B-Instruct)

SmolLM 1.7B Instruct Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
Research, Commercial applications
Primary Use Cases:
General knowledge questions, Creative writing, Basic Python programming
Limitations:
English only, Difficulty with arithmetic tasks, Difficulty with editing tasks, Complex reasoning limitations
Considerations:
Use as assistive tools rather than definitive sources of information.
Supported Languages 
en (English)
Training Details 
Data Sources:
Magpie-Align/Magpie-Pro-300K-Filtered, bigcode/self-oss-instruct-sc2-exec-filter-50k, teknium/OpenHermes-2.5, HuggingFaceTB/everyday-conversations-llama3.1-2k
Methodology:
Finetuning on publicly available datasets using DPO (Direct Preference Optimization) for certain models.
Training Time:
1 epoch for v0.2
Input Output 
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Using a temperature of 0.2 and top-p of 0.9 is suggested.
Release Notes 
Version:
v0.1
Notes:
Initial release with finetuning on the WebInstructSub dataset, combined with StarCoder2-Self-OSS-Instruct dataset, and DPO for certain models.
Version:
v0.2
Notes:
Changed finetuning mix to more suitable datasets. Improved performance on staying on topic and responding appropriately to standard prompts.
LLM NameSmolLM 1.7B Instruct
Repository 🤗https://huggingface.co/HuggingFaceTB/SmolLM-1.7B-Instruct 
Base Model(s)  HuggingFaceTB/SmolLM-1.7B   HuggingFaceTB/SmolLM-1.7B
Model Size1.7b
Required VRAM3.4 GB
Updated2024-12-22
MaintainerHuggingFaceTB
Model Typellama
Instruction-BasedYes
Model Files  3.4 GB
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.42.3
Tokenizer ClassGPT2Tokenizer
Padding Token<|im_end|>
Vocabulary Size49152
Torch Data Typebfloat16

Quantized Models of the SmolLM 1.7B Instruct

Model
Likes
Downloads
VRAM
SmolLM 1.7B Instruct GGUF41350 GB

Best Alternatives to SmolLM 1.7B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
SmolLM2 1.7B Instruct8K / 3.4 GB92671437
SmolTulu 1.7B Reinforced8K / 3.4 GB355
SmolLM2 Math IIO 1.7B Instruct8K / 3.4 GB1058
SmolLM2 1.7 Persona8K / 3.5 GB180
SmolLM2 1.7B Instruct8K / 3.4 GB9763
SmolLM2 1.7B Instruct8K / 3.4 GB1383
SmolLM 1.7B Instruct2K / 3.4 GB14810
SmolLM 1.7B2K / 3.4 GB8780
SmolLM 1.7B Instruct IFEval2K / 3.4 GB260
SmolLM2 1.7B Instruct Bnb 4bit8K / 1 GB14910
Note: green Score (e.g. "73.2") means that the model is better than HuggingFaceTB/SmolLM-1.7B-Instruct.

Rank the SmolLM 1.7B Instruct Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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  

What open-source LLMs or SLMs are you in search of? 40123 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217
  Support LLM Explorer