Pretrained Text-Generation Models Below 250M Parameters - a Felladrin Collection

Felladrin 's Collections

Frequently Used Spaces

Trained Models 🏋️

Pretrained Text-Generation Models Below 250M Parameters

Pretrained Text-Generation Models Below 250M Parameters

updated Dec 3, 2024

Great candidates for fine-tuning targeting Transformers.js, ordered by number of parameters.

M4-ai/TinyMistral-248M-v3

Text Generation • Updated Oct 20, 2024 • 187 • 5

Note License: apache-2.0
princeton-nlp/Sheared-Pythia-160m

Text Generation • Updated Dec 4, 2023 • 128 • 4

Note License: apache-2.0
JackFram/llama-160m

Text Generation • Updated Jan 4, 2024 • 223k • 34

Note License: apache-2.0
HuggingFaceTB/SmolLM-135M

Text Generation • Updated Aug 1, 2024 • 137k • 191

Note License: apache-2.0
amd/AMD-Llama-135m

Text Generation • Updated Oct 9, 2024 • 14.4k • 110

Note License: apache-2.0
BEE-spoke-data/smol_llama-101M-GQA

Text Generation • Updated Dec 25, 2023 • 3.99k • 28

Note License: apache-2.0
JackFram/llama-68m

Text Generation • Updated May 23, 2024 • 523k • 27

Note License: apache-2.0
Felladrin/Minueza-32M-Base

Text Generation • Updated Mar 10, 2024 • 407 • 15

Note License: apache-2.0
EleutherAI/pythia-31m

Text Generation • Updated Jul 26, 2023 • 18.6k • 5

Note License: apache-2.0