forum
Trying to backward through the graph a second time (or directly access saved tensors after they have
Test
Classification weights not loading
Open
Proper Support for RTX 5070 (5000 Series GPUs) - Unsloth
Avoiding overfitting
Unsloth issues with Medgemma 27B model (unsloth/medgemma-27b-text-it-bnb-4bit)
GRPO: No attribute "disable_adapter"
Open
Repeat all instructions
Dynamic 2.0 GGUF Quants
Open
Difference between google/gemma-3n-E4B-it and unsloth/gemma-3n-E4B-it
Fine Tuning Embedding with Unsloth
Open
Gpt-oss -20b- f16
DeepSeek V3.1 Help - any quant guide I can read
Open
The batch_size will be automatically reduced, and the training will restart from the first step.
Gemma 3 4b training not starting, help!
Open
Can I use Unsloth to fine-tune model for sequence classification?
Error when finetune with Orpheus notebook
Distilled-Deepseek R1 does not learn or even gets the scope of the wanted output.
Blackwell 5090 and RTX Pro - Working Dockerfile
Help regarding continious pretrain on gemma3 270M
Guidance on Choosing the Best Model for My Project
Open
Want to push the finetuned 4bit model to hugging face
Open
got-oss-4bit
ValueError: binary mode doesn't take an encoding argument
Link missing for Hyper-parameter tutorial
4-Bit Quant
Gemma finetuning train_on_responses_only Error
Crash on unsloth import in cluster environment (V100, no root)
How to set good parameter when fine tuning?
Open
Multi-GPU possible for SFT with Unsloth?
Solved
Llama 3.1:8b notebook error
Solved
gpt-oss-120b-F16 (gibberish answer)
Open
gpt-oss LoRA running
Constant OOM Problem
Open
Unsloth currently only works on NVIDIA GPUs and Intel GPUs
Open
UnSloth discord ai?
Horiza
Open
Hey I have a request
Open
"RuntimeError: shape '[-1, 8]' is invalid for input of size 7" while training with GRPO
Docker Desktop container for GPT‑OSS‑20B (F16) not utilizing my NVIDIA GPU
Fine-tuning Llama 3.2 - OOM Crash During Eval
Qwen3-4B-2507 talking crazy in Ollama
Open
load_in_4bit
How long will it take for GPT -oss 20b to come into 4-bit quantized
How to further finetune a model from a checkpoint
gpt-oss:20b not loading on Mac using transformers library from Huggingface
OpenAI OSS - Not formatting outputs correctly
Open
Noob looking to train an abliterated R1 model
Slow training speed
Open
save_pretrained_merged create a .cache folder