Fine-tuning open-source large and small instruct/chat language models (LLMs & SLMs) from the Hugging Face Model Hub using public datasets.
-
Updated
Aug 31, 2024 - Jupyter Notebook
#
smollm
Here are 16 public repositories matching this topic...
Build LLM-enabled FastAPI applications without build configuration.
nlp cli text-generation named-entity-recognition summarization llama ner fastapi huggingface llm langchain llamacpp langchain-python llama2 gemma-2b smollm
-
Updated
Jun 12, 2026 - Python
Kurtis is a fine-tuning, inference and evaluation tool built for SLMs (Small Language Models), such as Huggingface's SmolLM2.
machine-learning ai ml transformers question-answering mental-health mental-health-awareness slms huggingface-transformers llm llm-training llm-inference small-language-models huggingface-peft question-answering-model smollm smollm2 huggingface-smollm2
-
Updated
Mar 17, 2026 - Python
A fine-tuned version of SmolLM2-360M-Instruct-bnb-4bit specialized for parsing unstructured calendar event requests into structured JSON data.
json transformers llama fine-tuning calender-event huggingface huggingface-transformers huggingface-models text-to-json generative-ai instruction-tuning llamacpp finetuning-llms ollama finetuning-large-language-models small-language-models smollm smollm2 text2json text-2-json
-
Updated
Jun 21, 2025 - Python
Offline, on-device LLM browser for Android & iOS — React Native + llama.rn + SmolLM2 (Apache 2.0). 100% private: pages never leave your phone.
android open-source ios privacy react-native webview offline-first expo on-device-ai llm llama-cpp gguf mobile-browser smollm llama-rn
-
Updated
May 20, 2026 - TypeScript
SmolLM-360M distilled with Tree-of-Thoughts reasoning from GPT-4o, achieving competitive arithmetic reasoning with minimal compute
-
Updated
Dec 10, 2024 - Python
-
Updated
Feb 28, 2025 - Jupyter Notebook
A fun web app that generates clever, cringey, or romantic pickup lines using SmolLM-135M. Built with Next.js and TailwindCSS. Perfect for dating apps or breaking the ice!
-
Updated
Mar 25, 2025 - Python
Self-evaluation framework for LLM confidence calibration. Extracts True/False logits on TriviaQA; temperature scaling reduces ECE from 0.217→0.132 (Qwen-2.5-1.5B) without model modification.
nlp pytorch huggingface callibration temperature-scaling llm self-evaluation qwen smollm uncertainity-quantification
-
Updated
May 31, 2026 - Jupyter Notebook
⚙️ LoRA implementation for efficient SmolLM fine-tuning. Achieves comparable performance with only 0.24% trainable parameters.
nlp deep-learning pytorch transformer lora language-model fine-tuning peft huggingface low-rank-adaptation parameter-efficient-fine-tuning smollm
-
Updated
Jul 28, 2025 - Jupyter Notebook
Local speech-to-speech voice app for talking with cartoon characters
-
Updated
Jun 3, 2026 - JavaScript
Single-command LoRA SFT recipes for small open LLMs with reproducible before/after eval. CPU + GPU configs. SmolLM2-135M + SST-2: 460k trainable params (0.34% of base), +2 abs points in 5 min on a single CPU core.
-
Updated
Jun 10, 2026 - Python
Supervised Finetuning on SmolLM model using the smalltalk dataset.
-
Updated
Jul 25, 2025 - Jupyter Notebook
A 3B model that knows when it's unsure and spends compute only where it pays. Reproducible, on a laptop. Built on SmolLM3.
machine-learning calibration uncertainty-quantification mlx confidence-estimation selective-prediction llm small-language-models smollm test-time-compute
-
Updated
Jun 13, 2026 - Python
Improve this page
Add a description, image, and links to the smollm topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the smollm topic, visit your repo's landing page and select "manage topics."