📒 Fine-tuning Notebooks
Below are Colab notebooks, organized by model. You can also view all notebooks in our docs.
The notebooks run locally and feature data prep, training and inference. Read our fine-tuning guide.
Main Notebooks
| Model | Type | Notebook Link |
|---|
| Unsloth Studio | Chat UI |  |
| Gemma 4 (E2B) | Vision | |
| Qwen3.5 (4B) | Vision | |
| Qwen3.5 (2B) | Vision | |
| gpt-oss (20B) | Fine-tuning | |
| gpt-oss (20B) | GRPO | |
| Qwen3 (14B) | Conversational | |
| Qwen3-VL (8B) | Vision | |
| Qwen3-Embedding (0.6B) | Embeddings | |
| Qwen3: Advanced GRPO | GRPO | |
| Gemma 3 (4B) | Vision | |
| Gemma 3N (4B) | Audio | |
| embeddinggemma (300M) | Embeddings | |
| Mistral Ministral 3 (3B) | Vision | |
| Mistral v0.3 (7B) | Vision | |
| Llama 3.1 (8B) Alpaca | Alpaca | |
| Llama 3.2 (1B + 3B) | Conversational | |
| Phi-4 (14B) | Conversational | |
| Orpheus-TTS (3B) | TTS | |
Gemma 4 Notebooks
| Model | Type | Notebook Link |
|---|
| Gemma4 (E2B) | Sudoku (GRPO RL) | |
| Gemma4 (E2B) | Auto Kernel Creation (GRPO RL) | |
| Gemma4 (E2B) | 2048 Game (GRPO RL) | |
| Gemma4 (31B) | Vision | |
| Gemma4 (31B) | Conversational | |
| Gemma4 (E4B) | Vision | |
| Gemma4 (E4B) | Conversational | |
| Gemma4 (E4B) | Audio | |
| Gemma4 (26B A4B) | Vision | |
| Gemma4 (26B A4B) | Conversational | |
| Gemma4 (E2B) | Vision | |
| Gemma4 (E2B) | Conversational | |
| Gemma4 (E2B) | Audio | |
GRPO & Reinforcement Learning Notebooks
| Model | Type | Notebook Link |
|---|
| Llama3.1 (8B) | GSM8K Math + vLLM | |
| NeMo Gym Sudoku | Sudoku | |
| NeMo Gym Multi Environment | Multi Environment | |
| gpt oss BF16 (20B) | 2048 Game | |
| gpt oss (20B) | Minesweeper Game | |
| gpt oss (20B) | Auto Kernel Creation | |
| Qwen3 (8B) | DAPO Math + vLLM | |
| Llama3 (8B) | ORPO | |
| Openenv wordle | Wordle + vLLM | |
| Qwen2.5 VL (7B) | Vision Math + vLLM | |
| Zephyr (7B) | DPO | |
| (OpenEnv) gpt oss BF16 (20B) | 2048 Game | |
| gpt oss (20B) | Auto Kernel Creation | |
| gpt oss (20B) | 2048 Game | |
| (OpenEnv) gpt oss (20B) | 2048 Game | |
| (DGX Spark) gpt oss (20B) | 2048 Game | |
| (A100) gpt oss (20B) | Auto Kernel Creation | |
| Qwen2.5 (3B) | GSM8K Math + vLLM | |
| Llama3.2 (1B) | DAPO Math + vLLM | |
| Qwen3 VL (8B) | Vision Math | |
| Mistral v0.3 (7B) | GSM8K Math + vLLM | |
| Qwen3 5 (4B) | Vision Math | |
| Gemma3 (4B) | Vision Math | |
| Phi 4 (14B) | GSM8K Math + vLLM | |
| Gemma3 (1B) | GSM8K Math | |
| DeepSeek R1 0528 Qwen3 (8B) | DAPO Math + vLLM | |
| LFM2.5 (1.2B) | DAPO Math | |
| Gemma4 (E2B) | Sudoku | |
| Gemma4 (E2B) | Auto Kernel Creation | |
| Gemma4 (E2B) | 2048 Game | |
| Qwen3 (4B) | DAPO Math + vLLM | |
| Ministral3 (3B) | Sudoku | |
Tool Calling Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen2.5 Coder (1.5B) | Tool Calling | |
| FunctionGemma (270M) | Tool Calling | |
| FunctionGemma (270M) | Mobile Actions | |
| FunctionGemma (270M) | Inference | |
| FunctionGemma (270M) | Conversational | |
Text-to-Speech (TTS) Notebooks
| Model | Type | Notebook Link |
|---|
| Spark TTS (0.5B) | TTS | |
| Llasa TTS (1B) | TTS | |
| Orpheus (3B) | TTS | |
| Llasa TTS (3B) | TTS | |
| Sesame CSM (1B) | TTS | |
| Oute TTS (1B) | TTS | |
Vision (Multimodal) Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen2.5 VL (7B) | Vision Math + vLLM (GRPO RL) | |
| Qwen2.5 VL (7B) | Vision | |
| ERNIE 4 5 VL 28B A3B PT | Vision | |
| Qwen3 VL (8B) | Vision | |
| Qwen3 VL (8B) | Vision Math (GRPO RL) | |
| Qwen3 5 (4B) | Vision | |
| Qwen3 5 (4B) | Vision Math (GRPO RL) | |
| Gemma3 (4B) | Vision Math (GRPO RL) | |
| Qwen3 5 (0 8B) | Vision | |
| Gemma4 (31B) | Vision | |
| Qwen2 VL (7B) | Vision | |
| Gemma4 (E4B) | Vision | |
| Llama3.2 (11B) | Vision | |
| Qwen3 5 (2B) | Vision | |
| Gemma4 (26B A4B) | Vision | |
| Gemma4 (E2B) | Vision | |
| Pixtral (12B) | Vision | |
| LFM2.5 VL (1.6B) | Vision | |
| Ministral3 VL (3B) | Vision | |
| Gemma3 (4B) | Vision | |
| Gemma3N (4B) | Vision | |
Embedding Notebooks
| Model | Type | Notebook Link |
|---|
| ModernBert | Classification | |
| Qwen3 Embedding (4B) | Embeddings | |
| Qwen3 Embedding (0 6B) | Embeddings | |
| EmbeddingGemma (300M) | Embeddings | |
| ModernBERT (Large) | Classification | |
| BGE M3 | Embeddings | |
| All MiniLM L6 v2 | Embeddings | |
Speech-to-Text (STT) Notebooks
| Model | Type | Notebook Link |
|---|
| Whisper (Large) | Fine Tuning | |
OCR Notebooks
| Model | Type | Notebook Link |
|---|
| Deepseek OCR (3B) | Fine Tuning | |
| Deepseek OCR (3B) | Evaluation | |
| Deepseek OCR (3B) | Eval | |
| Deepseek OCR 2 (3B) | Fine Tuning | |
| Paddle OCR (1B) | Vision | |
BERT Notebooks
| Model | Type | Notebook Link |
|---|
| ModernBert | Classification | |
| ModernBERT (Large) | Classification | |
Deepseek Notebooks
| Model | Type | Notebook Link |
|---|
| Deepseek OCR (3B) | Fine Tuning | |
| Deepseek OCR (3B) | Evaluation | |
| Deepseek OCR (3B) | Eval | |
| Deepseek OCR 2 (3B) | Fine Tuning | |
ERNIE Notebooks
| Model | Type | Notebook Link |
|---|
| ERNIE 4 5 VL 28B A3B PT | Vision | |
| ERNIE 4 5 21B A3B PT | Conversational | |
GLM Notebooks
| Model | Type | Notebook Link |
|---|
| (A100) GLM Flash(80GB) | Conversational | |
GPT-OSS Notebooks
| Model | Type | Notebook Link |
|---|
| gpt oss MXFP4 (20B) | Inference | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss BNB (20B) | Inference | |
| (A100) gpt oss (120B) | Fine Tuning | |
Gemma Notebooks
| Model | Type | Notebook Link |
|---|
| Gemma3 (4B) | Conversational | |
| (A100) Gemma3 (27B) | Conversational | |
| Gemma3 (270M) | Phone Deployment | |
| Gemma3 (270M) | Conversational | |
| Gemma3N (4B) | Multimodal | |
| Gemma3N (4B) | Audio | |
| Gemma3N (2B) | Inference | |
| Gemma3 (4B) | Vision | |
| Gemma3N (4B) | Vision | |
| EmbeddingGemma (300M) | Embeddings | |
| Gemma2 (9B) | Alpaca | |
| Gemma2 (2B) | Alpaca | |
| CodeGemma (7B) | Conversational | |
| DiffusionGemma (26B A4B) | Sudoku | |
Granite Notebooks
| Model | Type | Notebook Link |
|---|
| Granite4.0 (3B) | Conversational | |
| Granite4.0 (350M) | Conversational | |
Hybrid Attention Notebooks
| Model | Type | Notebook Link |
|---|
| LFM2.5 (1.2B) | Conversational | |
| Liquid LFM2 (1.2B) | Conversational | |
| Liquid LFM2 | Conversational | |
| LFM2.5 VL (1.6B) | Vision | |
| LFM2.5 (1.2B) | Translation | |
| Falcon H1 (0.5B) | Alpaca | |
| Falcon H1 | Alpaca | |
Llama Notebooks
| Model | Type | Notebook Link |
|---|
| Llama3.1 (8B) | Inference | |
| Llama3 (8B) | Ollama | |
| Llama3 (8B) | Alpaca | |
| Llama3.2 (1B) | RAFT | |
| Llama3.2 (1B and 3B) | Conversational | |
| Llama3.1 (8B) | Alpaca | |
| (A100) Llama3.3 (70B) | Conversational | |
| Llama3.2 (11B) | Vision | |
| Llama3 (8B) | Conversational | |
| TinyLlama (1.1B) | Alpaca | |
Mistral Notebooks
| Model | Type | Notebook Link |
|---|
| Mistral v0.3 (7B) | Conversational | |
| Magistral (24B) | Reasoning Conversational | |
| Pixtral (12B) | Vision | |
| Mistral Small (22B) | Alpaca | |
| Ministral3 VL (3B) | Vision | |
| Mistral v0.3 (7B) | Alpaca | |
| Mistral Nemo (12B) | Alpaca | |
Nemotron Notebooks
| Model | Type | Notebook Link |
|---|
| (A100) Nemotron Nano 3 30B A3B | Conversational | |
| (A100) Nemotron 3 Nano 30B A3B | Conversational | |
Paddle Notebooks
| Model | Type | Notebook Link |
|---|
| Paddle OCR (1B) | Vision | |
Phi Notebooks
| Model | Type | Notebook Link |
|---|
| Phi 4 | Conversational | |
| Phi 3.5 Mini | Conversational | |
| Phi 3 Medium | Conversational | |
Qwen Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen3 (0.6B) | Reasoning Conversational | |
| Qwen3 (0 6B) | Phone Deployment | |
| Qwen3 (4B) | QAT | |
| Qwen3 (4B) | Conversational | |
| Qwen2.5 VL (7B) | Vision | |
| Qwen3 VL (8B) | Vision | |
| Qwen3 5 MoE | MoE | |
| (A100) Qwen 3 5 27B(80GB) | Conversational | |
| (A100) Qwen3 (32B) | Reasoning Conversational | |
| Qwen3 5 (4B) | Vision | |
| Qwen3 (14B) | Reasoning Conversational | |
| Qwen3 (14B) | Conversational | |
| Qwen3 5 (0 8B) | Vision | |
| Qwen2 VL (7B) | Vision | |
| Qwen3 (4B) | Thinking | |
| Qwen3 MoE | MoE | |
| Qwen3 5 (2B) | Vision | |
| Qwen2.5 (7B) | Alpaca | |
| Qwen2.5 Coder (14B) | Conversational | |
| Qwen3 6 MoE | MoE | |
| Qwen3 Embedding (4B) | Embeddings | |
| Qwen3 Embedding (0 6B) | Embeddings | |
| Qwen3 (14B) | Alpaca | |
| Qwen2 (7B) | Alpaca | |
| TinyQwen3 MoE | MoE | |
Text Completion / Continued Pretraining Notebooks
| Model | Type | Notebook Link |
|---|
| LFM2.5 (1.2B) | Text Completion | |
| Mistral v0.3 (7B) | CPT | |
| Mistral (7B) | Text Completion | |
Specific use-case Notebooks
| Usecase | Model | Notebook Link |
|---|
| Text Classification | Llama 3.1 (8B) | |
| Tool Calling | Qwen2.5-Coder (1.5B) | |
| Multiple Datasets | | |
| KTO | Qwen2.5-Instruct (1.5B) | |
| Inference Chat UI | LLaMa 3.2 Vision | |
| Conversational | LLaMa 3.2 (1B and 3B) | |
| ChatML | Mistral (7B) | |
| Text Completion | Mistral (7B) | |
Other Notebooks
| Model | Type | Notebook Link |
|---|
| CodeForces CoT Reasoning | | |
| Synthetic Data Hackathon | Synthetic Data | |
| Unsloth | Studio | |
📒 Kaggle Notebooks
Click for all our Kaggle notebooks categorized by model:
GRPO & Reinforcement Learning Notebooks
| Model | Type | Notebook Link |
|---|
| Llama3.1 (8B) | GSM8K Math + vLLM |  |
| gpt oss (20B) | Minesweeper Game | |
| gpt oss (20B) | Auto Kernel Creation | |
| Qwen3 (8B) | DAPO Math + vLLM | |
| Llama3 (8B) | ORPO | |
| Qwen2.5 VL (7B) | Vision Math + vLLM | |
| Ministral3 (3B) | Sudoku | |
| Zephyr (7B) | DPO | |
| gpt oss (20B) | Auto Kernel Creation | |
| (A100) gpt oss (20B) | Auto Kernel Creation | |
| Qwen2.5 (3B) | GSM8K Math + vLLM | |
| Llama3.2 (1B) | DAPO Math + vLLM | |
| Qwen3 VL (8B) | Vision Math | |
| Mistral v0.3 (7B) | GSM8K Math + vLLM | |
| Gemma3 (4B) | Vision Math | |
| Phi 4 (14B) | GSM8K Math + vLLM | |
| Gemma3 (1B) | GSM8K Math | |
| DeepSeek R1 0528 Qwen3 (8B) | DAPO Math + vLLM | |
| Qwen3 (4B) | DAPO Math + vLLM | |
Tool Calling Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen2.5 Coder (1.5B) | Tool Calling | |
Text-to-Speech (TTS) Notebooks
| Model | Type | Notebook Link |
|---|
| Spark TTS (0.5B) | TTS | |
| Llasa TTS (1B) | TTS | |
| Orpheus (3B) | TTS | |
| Llasa TTS (3B) | TTS | |
| Sesame CSM (1B) | TTS | |
| Oute TTS (1B) | TTS | |
Vision (Multimodal) Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen2.5 VL (7B) | Vision Math + vLLM (GRPO RL) | |
| Qwen2.5 VL (7B) | Vision | |
| ERNIE 4 5 VL 28B A3B PT | Vision | |
| Qwen3 VL (8B) | Vision | |
| Qwen3 VL (8B) | Vision Math (GRPO RL) | |
| Gemma3 (4B) | Vision Math (GRPO RL) | |
| Qwen2 VL (7B) | Vision | |
| Llama3.2 (11B) | Vision | |
| Pixtral (12B) | Vision | |
| Ministral3 VL (3B) | Vision | |
| Gemma3 (4B) | Vision | |
| Gemma3N (4B) | Vision | |
Embedding Notebooks
| Model | Type | Notebook Link |
|---|
| ModernBert | Classification | |
| Qwen3 Embedding (4B) | Embeddings | |
| Qwen3 Embedding (0 6B) | Embeddings | |
| EmbeddingGemma (300M) | Embeddings | |
| ModernBERT (Large) | Classification | |
| BGE M3 | Embeddings | |
| All MiniLM L6 v2 | Embeddings | |
Speech-to-Text (STT) Notebooks
| Model | Type | Notebook Link |
|---|
| Whisper (Large) | Fine Tuning | |
OCR Notebooks
| Model | Type | Notebook Link |
|---|
| Deepseek OCR (3B) | Fine Tuning | |
| Deepseek OCR (3B) | Evaluation | |
| Deepseek OCR (3B) | Eval | |
| Deepseek OCR 2 (3B) | Fine Tuning | |
| Paddle OCR (1B) | Vision | |
BERT Notebooks
| Model | Type | Notebook Link |
|---|
| ModernBert | Classification | |
| ModernBERT (Large) | Classification | |
Deepseek Notebooks
| Model | Type | Notebook Link |
|---|
| Deepseek OCR (3B) | Fine Tuning | |
| Deepseek OCR (3B) | Evaluation | |
| Deepseek OCR (3B) | Eval | |
| Deepseek OCR 2 (3B) | Fine Tuning | |
ERNIE Notebooks
| Model | Type | Notebook Link |
|---|
| ERNIE 4 5 VL 28B A3B PT | Vision | |
| ERNIE 4 5 21B A3B PT | Conversational | |
GPT-OSS Notebooks
| Model | Type | Notebook Link |
|---|
| gpt oss MXFP4 (20B) | Inference | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss BNB (20B) | Inference | |
| (A100) gpt oss (120B) | Fine Tuning | |
Gemma Notebooks
| Model | Type | Notebook Link |
|---|
| Gemma3 (4B) | Conversational | |
| (A100) Gemma3 (27B) | Conversational | |
| Gemma3 (270M) | Conversational | |
| Gemma3N (4B) | Multimodal | |
| Gemma3N (4B) | Audio | |
| Gemma3N (2B) | Inference | |
| Gemma3 (4B) | Vision | |
| Gemma3N (4B) | Vision | |
| EmbeddingGemma (300M) | Embeddings | |
| Gemma2 (9B) | Alpaca | |
| Gemma2 (2B) | Alpaca | |
| CodeGemma (7B) | Conversational | |
| DiffusionGemma (26B A4B) | Sudoku | |
Granite Notebooks
| Model | Type | Notebook Link |
|---|
| Granite4.0 (3B) | Conversational | |
| Granite4.0 (350M) | Conversational | |
Hybrid Attention Notebooks
| Model | Type | Notebook Link |
|---|
| Liquid LFM2 (1.2B) | Conversational | |
| Falcon H1 (0.5B) | Alpaca | |
Llama Notebooks
| Model | Type | Notebook Link |
|---|
| Llama3.1 (8B) | Inference | |
| Llama3 (8B) | Ollama | |
| Llama3 (8B) | Alpaca | |
| Llama3.2 (1B) | RAFT | |
| Llama3.2 (1B and 3B) | Conversational | |
| Llama3.1 (8B) | Alpaca | |
| (A100) Llama3.3 (70B) | Conversational | |
| Llama3.2 (11B) | Vision | |
| Llama3 (8B) | Conversational | |
| TinyLlama (1.1B) | Alpaca | |
Mistral Notebooks
| Model | Type | Notebook Link |
|---|
| Mistral v0.3 (7B) | Conversational | |
| Magistral (24B) | Reasoning Conversational | |
| Pixtral (12B) | Vision | |
| Mistral Small (22B) | Alpaca | |
| Ministral3 VL (3B) | Vision | |
| Mistral v0.3 (7B) | Alpaca | |
| Mistral Nemo (12B) | Alpaca | |
Nemotron Notebooks
| Model | Type | Notebook Link |
|---|
| (A100) Nemotron Nano 3 30B A3B | Conversational | |
| (A100) Nemotron 3 Nano 30B A3B | Conversational | |
Paddle Notebooks
| Model | Type | Notebook Link |
|---|
| Paddle OCR (1B) | Vision | |
Phi Notebooks
| Model | Type | Notebook Link |
|---|
| Phi 4 | Conversational | |
| Phi 3.5 Mini | Conversational | |
| Phi 3 Medium | Conversational | |
Qwen Notebooks
| Model | Type | Notebook Link |
|---|
| Qwen3 (4B) | QAT | |
| Qwen3 (4B) | Conversational | |
| Qwen2.5 VL (7B) | Vision | |
| Qwen3 VL (8B) | Vision | |
| (A100) Qwen3 (32B) | Reasoning Conversational | |
| Qwen3 (14B) | Reasoning Conversational | |
| Qwen3 (14B) | Conversational | |
| Qwen2 VL (7B) | Vision | |
| Qwen3 (4B) | Thinking | |
| Qwen2.5 (7B) | Alpaca | |
| Qwen2.5 Coder (14B) | Conversational | |
| Qwen3 Embedding (4B) | Embeddings | |
| Qwen3 Embedding (0 6B) | Embeddings | |
| Qwen3 (14B) | Alpaca | |
| Qwen2 (7B) | Alpaca | |
Text Completion / Continued Pretraining Notebooks
| Model | Type | Notebook Link |
|---|
| Mistral v0.3 (7B) | CPT | |
| Mistral (7B) | Text Completion | |
Other Notebooks
| Model | Type | Notebook Link |
|---|
| CodeForces CoT Reasoning | | |
| Unsloth | Studio | |
🐧 AMD Notebooks
These notebooks target AMD ROCm GPUs and are not available in Colab. View / download them directly from GitHub:
| Model | Type | Notebook |
|---|
| Unsloth Studio | Chat UI | GitHub |
| Gemma4 (E2B) | Vision | GitHub |
| Qwen3 5 (4B) | Vision | GitHub |
| Qwen3 5 (2B) | Vision | GitHub |
| gpt oss (20B) | Fine Tuning | GitHub |
| gpt oss (20B) | Auto Kernel Creation | GitHub |
Click for all our AMD ROCm notebooks:
| Model | Type | Notebook |
|---|
| Qwen3 (0 6B) | Phone Deployment | GitHub |
| Qwen3 (0.6B) | Reasoning Conversational | GitHub |
| Llama3.1 (8B) | Inference | GitHub |
| Llama3.1 (8B) | GSM8K Math + vLLM | GitHub |
| NeMo Gym Sudoku | Sudoku | GitHub |
| NeMo Gym Multi Environment | Multi Environment | GitHub |
| Whisper (Large) | Fine Tuning | GitHub |
| gpt oss MXFP4 (20B) | Inference | GitHub |
| gpt oss BNB (20B) | Inference | GitHub |
| gpt oss BF16 (20B) | 2048 Game | GitHub |
| gpt oss (20B) | 2048 Game | GitHub |
| gpt oss (20B) | Minesweeper Game | GitHub |
| gpt oss (20B) | Auto Kernel Creation | GitHub |
| gpt oss (20B) | Fine Tuning | GitHub |
| (OpenEnv) gpt oss BF16 (20B) | 2048 Game | GitHub |
| (OpenEnv) gpt oss (20B) | 2048 Game | GitHub |
| (DGX Spark) gpt oss (20B) | 2048 Game | GitHub |
| Spark TTS (0.5B) | TTS | GitHub |
| Qwen3 (8B) | DAPO Math + vLLM | GitHub |
| Llama3 (8B) | Ollama | GitHub |
| Llama3 (8B) | ORPO | GitHub |
| Llama3 (8B) | Alpaca | GitHub |
| Openenv wordle | Wordle + vLLM | GitHub |
| gpt oss (20B) | Auto Kernel Creation | GitHub |
| gpt oss (120B) | Fine Tuning | GitHub |
| Qwen2.5 (3B) | GSM8K Math + vLLM | GitHub |
| ModernBert | Classification | GitHub |
| Qwen3 (4B) | QAT | GitHub |
| Qwen3 (4B) | Conversational | GitHub |
| Qwen2.5 VL (7B) | Vision Math + vLLM | GitHub |
| Qwen2.5 VL (7B) | Vision | GitHub |
| Llasa TTS (1B) | TTS | GitHub |
| Llama3.2 (1B) | DAPO Math + vLLM | GitHub |
| Llama3.2 (1B) | RAFT | GitHub |
| Deepseek OCR (3B) | Fine Tuning | GitHub |
| Deepseek OCR (3B) | Evaluation | GitHub |
| Deepseek OCR (3B) | Eval | GitHub |
| Paddle OCR (1B) | Vision | GitHub |
| ERNIE 4 5 VL 28B A3B PT | Vision | GitHub |
| Deepseek OCR 2 (3B) | Fine Tuning | GitHub |
| Qwen3 VL (8B) | Vision | GitHub |
| Qwen3 VL (8B) | Vision Math | GitHub |
| Mistral v0.3 (7B) | GSM8K Math + vLLM | GitHub |
| Mistral v0.3 (7B) | Conversational | GitHub |
| Qwen3 5 MoE | MoE | GitHub |
| Orpheus (3B) | TTS | GitHub |
| Llasa TTS (3B) | TTS | GitHub |
| Meta Synthetic Data Llama3.1 (8B) | GRPO | GitHub |
| Meta Synthetic Data Llama3 2 (3B) | GRPO | GitHub |
| Llama3.2 (1B and 3B) | Conversational | GitHub |
| Qwen 3 5 27B(80GB) | Conversational | GitHub |
| Qwen3 (32B) | Reasoning Conversational | GitHub |
| Llama3.1 (8B) | Alpaca | GitHub |
| Qwen3 5 (4B) | Vision Math | GitHub |
| Qwen3 (14B) | Conversational | GitHub |
| Qwen3 (14B) | Reasoning Conversational | GitHub |
| CodeForces CoT Reasoning | | GitHub |
| Llama3.3 (70B) | Conversational | GitHub |
| Synthetic Data Hackathon | Synthetic Data | GitHub |
| Gemma3 (4B) | Conversational | GitHub |
| Gemma3 (4B) | Vision Math | GitHub |
| Phi 4 (14B) | GSM8K Math + vLLM | GitHub |
| Phi 4 | Conversational | GitHub |
| Gemma3 (27B) | Conversational | GitHub |
| Qwen3 5 (0 8B) | Vision | GitHub |
| GLM Flash(80GB) | Conversational | GitHub |
| Sesame CSM (1B) | TTS | GitHub |
| Gemma4 (31B) | Conversational | GitHub |
| Gemma4 (31B) | Vision | GitHub |
| Qwen2 VL (7B) | Vision | GitHub |
| Qwen3 (4B) | Thinking | GitHub |
| Qwen3 MoE | MoE | GitHub |
| Gemma3 (1B) | GSM8K Math | GitHub |
| Nemotron Nano 3 30B A3B | Conversational | GitHub |
| Nemotron 3 Nano 30B A3B | Conversational | GitHub |
| Gemma4 (E4B) | Conversational | GitHub |
| Gemma4 (E4B) | Vision | GitHub |
| Gemma4 (E4B) | Audio | GitHub |
| Llama3.2 (11B) | Vision | GitHub |
| Phi 3.5 Mini | Conversational | GitHub |
| Gemma4 (26B A4B) | Conversational | GitHub |
| Gemma4 (26B A4B) | Vision | GitHub |
| Magistral (24B) | Reasoning Conversational | GitHub |
| Qwen2.5 (7B) | Alpaca | GitHub |
| DeepSeek R1 0528 Qwen3 (8B) | DAPO Math + vLLM | GitHub |
| Qwen2.5 Coder (1.5B) | Tool Calling | GitHub |
| Gemma3 (270M) | Conversational | GitHub |
| Gemma3 (270M) | Phone Deployment | GitHub |
| Qwen2.5 Coder (14B) | Conversational | GitHub |
| FunctionGemma (270M) | Conversational | GitHub |
| FunctionGemma (270M) | Inference | GitHub |
| FunctionGemma (270M) | Tool Calling | GitHub |
| FunctionGemma (270M) | Mobile Actions | GitHub |
| Gemma3N (4B) | Multimodal | GitHub |
| Gemma3N (4B) | Audio | GitHub |
| Gemma3N (2B) | Inference | GitHub |
| LFM2.5 (1.2B) | DAPO Math | GitHub |
| LFM2.5 (1.2B) | Conversational | GitHub |
| Gemma4 (E2B) | Conversational | GitHub |
| Gemma4 (E2B) | Sudoku | GitHub |
| Gemma4 (E2B) | 2048 Game | GitHub |
| Gemma4 (E2B) | Auto Kernel Creation | GitHub |
| Gemma4 (E2B) | Audio | GitHub |
| Qwen3 6 MoE | MoE | GitHub |
| Qwen3 Embedding (4B) | Embeddings | GitHub |
| Qwen3 (4B) | DAPO Math + vLLM | GitHub |
| Pixtral (12B) | Vision | GitHub |
| Qwen3 Embedding (0 6B) | Embeddings | GitHub |
| Mistral Small (22B) | Alpaca | GitHub |
| Liquid LFM2 (1.2B) | Conversational | GitHub |
| Liquid LFM2 | Conversational | GitHub |
| LFM2.5 VL (1.6B) | Vision | GitHub |
| Ministral3 VL (3B) | Vision | GitHub |
| Ministral3 (3B) | Sudoku | GitHub |
| Gemma3 (4B) | Vision | GitHub |
| Oute TTS (1B) | TTS | GitHub |
| Llama3 (8B) | Conversational | GitHub |
| ERNIE 4 5 21B A3B PT | Conversational | GitHub |
| Granite4.0 (3B) | Conversational | GitHub |
| Qwen3 (14B) | Alpaca | GitHub |
| LFM2.5 (1.2B) | Translation | GitHub |
| LFM2.5 (1.2B) | Text Completion | GitHub |
| Gemma3N (4B) | Vision | GitHub |
| Granite4.0 (350M) | Conversational | GitHub |
| TinyLlama (1.1B) | Alpaca | GitHub |
| Falcon H1 (0.5B) | Alpaca | GitHub |
| Falcon H1 | Alpaca | GitHub |
| Phi 3 Medium | Conversational | GitHub |
| EmbeddingGemma (300M) | Embeddings | GitHub |
| Gemma2 (9B) | Alpaca | GitHub |
| Gemma2 (2B) | Alpaca | GitHub |
| Mistral v0.3 (7B) | CPT | GitHub |
| Mistral v0.3 (7B) | Alpaca | GitHub |
| Mistral (7B) | Text Completion | GitHub |
| Qwen2 (7B) | Alpaca | GitHub |
| Zephyr (7B) | DPO | GitHub |
| ModernBERT (Large) | Classification | GitHub |
| Mistral Nemo (12B) | Alpaca | GitHub |
| CodeGemma (7B) | Conversational | GitHub |
| BGE M3 | Embeddings | GitHub |
| TinyQwen3 MoE | MoE | GitHub |
| All MiniLM L6 v2 | Embeddings | GitHub |
🍃 Molab Notebooks
Run any of these on molab, Marimo's hosted GPU notebooks. They're reactive: change a value in one cell, the cells below recompute on their own.
| Model | Type | Notebook |
|---|
| Unsloth Studio | Chat UI |  |
| Gemma4 (E2B) | Vision | |
| Qwen3 5 (4B) | Vision | |
| Qwen3 5 (2B) | Vision | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss (20B) | Auto Kernel Creation | |
Click for all our molab notebooks:
| Model | Type | Notebook |
|---|
| All MiniLM L6 v2 | Embeddings | |
| BGE M3 | Embeddings | |
| CodeForces CoT Reasoning | | |
| CodeGemma (7B) | Conversational | |
| Deepseek OCR (3B) | Fine Tuning | |
| Deepseek OCR (3B) | Eval | |
| Deepseek OCR (3B) | Evaluation | |
| Deepseek OCR 2 (3B) | Fine Tuning | |
| ERNIE 4 5 21B A3B PT | Conversational | |
| ERNIE 4 5 VL 28B A3B PT | Vision | |
| EmbeddingGemma (300M) | Embeddings | |
| Falcon H1 | Alpaca | |
| Falcon H1 (0.5B) | Alpaca | |
| FunctionGemma (270M) | Conversational | |
| FunctionGemma (270M) | Inference | |
| FunctionGemma (270M) | Mobile Actions | |
| FunctionGemma (270M) | Tool Calling | |
| GLM Flash(80GB) | Conversational | |
| gpt oss BNB (20B) | Inference | |
| gpt oss MXFP4 (20B) | Inference | |
| Gemma2 (2B) | Alpaca | |
| Gemma2 (9B) | Alpaca | |
| Gemma3N (2B) | Inference | |
| Gemma3N (4B) | Audio | |
| Gemma3N (4B) | Multimodal | |
| Gemma3N (4B) | Vision | |
| Gemma3 (270M) | Conversational | |
| Gemma3 (270M) | Phone Deployment | |
| Gemma3 (27B) | Conversational | |
| Gemma3 (4B) | Conversational | |
| Gemma3 (4B) | Vision | |
| Gemma4 (12B) | Audio | |
| Gemma4 (12B) | | |
| Gemma4 (12B) | Vision | |
| Gemma4 (26B A4B) | Conversational | |
| Gemma4 (26B A4B) | Vision | |
| Gemma4 (31B) | Conversational | |
| Gemma4 (31B) | Vision | |
| Gemma4 (E2B) | Audio | |
| Gemma4 (E2B) | Conversational | |
| Gemma4 (E2B) | 2048 Game | |
| Gemma4 (E2B) | Sudoku | |
| Gemma4 (E4B) | Audio | |
| Gemma4 (E4B) | Conversational | |
| Gemma4 (E4B) | Vision | |
| Granite4.0 (3B) | Conversational | |
| Granite4.0 (350M) | Conversational | |
| LFM2.5 (1.2B) | Conversational | |
| LFM2.5 (1.2B) | Text Completion | |
| LFM2.5 (1.2B) | Translation | |
| LFM2.5 VL (1.6B) | Vision | |
| Liquid LFM2 | Conversational | |
| Liquid LFM2 (1.2B) | Conversational | |
| Llama3.1 (8B) | Alpaca | |
| Llama3.1 (8B) | Inference | |
| Llama3.2 (11B) | Vision | |
| Llama3.2 (1B) | RAFT | |
| Llama3.2 (1B and 3B) | Conversational | |
| Llama3.3 (70B) | Conversational | |
| Llama3 (8B) | Alpaca | |
| Llama3 (8B) | Conversational | |
| Llama3 (8B) | ORPO | |
| Llama3 (8B) | Ollama | |
| Llasa TTS (1B) | TTS | |
| Llasa TTS (3B) | TTS | |
| Magistral (24B) | Reasoning Conversational | |
| Ministral3 (3B) | Sudoku | |
| Ministral3 VL (3B) | Vision | |
| Mistral (7B) | Text Completion | |
| Mistral Nemo (12B) | Alpaca | |
| Mistral Small (22B) | Alpaca | |
| Mistral v0.3 (7B) | Alpaca | |
| Mistral v0.3 (7B) | CPT | |
| Mistral v0.3 (7B) | Conversational | |
| ModernBert | Classification | |
| NeMo Gym Multi Environment | Multi Environment | |
| NeMo Gym Sudoku | Sudoku | |
| Nemotron 3 Nano 30B A3B | Conversational | |
| Nemotron Nano 3 30B A3B | Conversational | |
| (OpenEnv) gpt oss (20B) | 2048 Game | |
| (OpenEnv) gpt oss BF16 (20B) | 2048 Game | |
| Openenv wordle | Wordle + vLLM | |
| Orpheus (3B) | TTS | |
| Oute TTS (1B) | TTS | |
| Paddle OCR (1B) | Vision | |
| Phi 3.5 Mini | Conversational | |
| Phi 3 Medium | Conversational | |
| Phi 4 | Conversational | |
| Pixtral (12B) | Vision | |
| Qwen2.5 (7B) | Alpaca | |
| Qwen2.5 Coder (1.5B) | Tool Calling | |
| Qwen2.5 Coder (14B) | Conversational | |
| Qwen2.5 VL (7B) | Vision | |
| Qwen2 (7B) | Alpaca | |
| Qwen2 VL (7B) | Vision | |
| Qwen3 (0.6B) | Reasoning Conversational | |
| Qwen3 (0 6B) | Phone Deployment | |
| Qwen3 (14B) | Conversational | |
| Qwen3 (14B) | Alpaca | |
| Qwen3 (14B) | Reasoning Conversational | |
| Qwen3 (32B) | Reasoning Conversational | |
| Qwen3 (4B) | Conversational | |
| Qwen3 (4B) | Thinking | |
| Qwen3 (4B) | QAT | |
| Qwen3 5 (0 8B) | Vision | |
| Qwen3 5 (4B) | Vision Math (GRPO RL) | |
| Qwen3 5 MoE | MoE | |
| Qwen3 6 MoE | MoE | |
| Qwen3 Embedding (0 6B) | Embeddings | |
| Qwen3 Embedding (4B) | Embeddings | |
| Qwen3 MoE | MoE | |
| Qwen3 VL (8B) | Vision | |
| Qwen3 VL (8B) | Vision Math (GRPO RL) | |
| Qwen 3 5 27B(80GB) | Conversational | |
| Sesame CSM (1B) | TTS | |
| Spark TTS (0.5B) | TTS | |
| Synthetic Data Hackathon | Synthetic Data | |
| TinyLlama (1.1B) | Alpaca | |
| TinyQwen3 MoE | MoE | |
| Whisper (Large) | Fine Tuning | |
| Zephyr (7B) | DPO | |
| ModernBERT (Large) | Classification | |
| gpt oss (120B) | Fine Tuning | |
| (A100) gpt oss (20B) | Auto Kernel Creation | |
| gpt oss (20B) | Fine Tuning | |
| gpt oss (20B) | Auto Kernel Creation | |
| gpt oss (20B) | 2048 Game | |
| gpt oss BF16 (20B) | 2048 Game | |
| (DGX Spark) gpt oss (20B) | 2048 Game | |
| gpt oss (20B) | Minesweeper Game | |
✨ Contributing to Notebooks
If you'd like to contribute to our notebooks, here's a guide to get you started:
- Find the Template: We've provided a template notebook called
Template_Notebook.ipynb in the root directory of this project. This template contains the basic structure and formatting guidelines for all notebooks in this collection.
- Create Your Notebook:
- Make a copy of
Template_Notebook.ipynb.
- Rename the copied file to follow this naming convention:
- LLM Notebooks:
<Model Name>-<Type>.ipynb (e.g., Mistral_v0.3_(7B)-Alpaca.ipynb)
- Vision Notebooks:
<Model Name>-Vision.ipynb (e.g., Llava_v1.6_(7B)-Vision.ipynb)
- Example of
<Type>: Alpaca, Conversational, CPT, DPO, ORPO, Text_Completion, CSV, Inference, Unsloth_Studio
- Place in
original_template: Once your notebook is ready, move it to the original_template directory.
- Update Notebooks: Run the following command in your terminal:
python update_all_notebooks.py
- Copy your notebook from
original_template to the notebooks directory.
- Update the notebook's internal sections (like Installation, News) to ensure consistency.
- Add your notebook to the appropriate list in this
README.md file.
- Create a Pull Request: After that, just create a pull request (PR) to merge your changes, making it available for everyone!
- We appreciate your contributions and look forward to reviewing your notebooks!
