LLM EXPLORER 59,420 MODELS INDEXED

SmollerLM 20M Instruct Pruned Base PostPruneRetraining by mehmetkeremturkcan

By mehmetkeremturkcan · 5 downloads
HF Repo ↗ Copy URL

SmollerLM 20M Instruct Pruned Base PostPruneRetraining is an open-source language model by mehmetkeremturkcan. Features: 20m LLM, Context: 8K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.16.

  Instruct   Llama   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/mehmetkeremturkcan/SmollerLM-20M-Instruct-Pruned-Base-PostPruneRetraining 

SmollerLM 20M Instruct Pruned Base PostPruneRetraining Parameters and Internals

LLM NameSmollerLM 20M Instruct Pruned Base PostPruneRetraining
Repository 🤗https://huggingface.co/mehmetkeremturkcan/SmollerLM-20M-Instruct-Pruned-Base-PostPruneRetraining 
Model Size20m
Required VRAM0 GB
Updated2026-06-29
Maintainermehmetkeremturkcan
Model Typellama
Instruction-BasedYes
Model Files  0.0 GB
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length8192
Model Max Length8192
Transformers Version4.48.2
Tokenizer ClassGPT2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size49152
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to SmollerLM 20M Instruct Pruned Base PostPruneRetraining

Best Alternatives
Context / RAM
Downloads
Likes
EAGLE3 GPT Oss 20B128K / 0.7 GB5070
Bagel 20B V04 Llama32K / 39.6 GB67
Bagel DPO 20B V04 Llama32K / 39.6 GB73
...lm2 1.7B Distilled GPT Oss 20B8K / 0 GB353
...lerLM 20M Instruct Pruned Sft58K / 0.1 GB120
Rose 20B4K / 39.9 GB11446
...I Llama2 Ko En Instruct 20B V12K / 40.4 GB50
Rose 20B GPTQ4K / 10.5 GB43
Rose 20B AWQ4K / 10.9 GB11
Note: green Score (e.g. "73.2") means that the model is better than mehmetkeremturkcan/SmollerLM-20M-Instruct-Pruned-Base-PostPruneRetraining.