forum

1495 threads · Page 17 of 30

Full-Fine tuning over unsloth 6 messages
Problems installing Unsloth 47 messages
Gemma3 4B vision in react native 2 messages
Audio/Speech Support 3 messages
Problem merging Lora with based model. 19 messages
General Advice for a Low-Resource Language 3 messages
Open
Fine tuning generates repetitive output 6 messages
Evaluation & custom compute_metrics can't produce coherent output 2 messages
Open
Multi-GPU / Unsloth Pro 5 messages
Phi_4_(14B)_GRPO.ipynb 18 messages
GRPO+VIsion Models 24 messages
DeepSeek-V3-0324-UD coding errors in output for Heptagon Test 8 messages
AttributeError: 'Gemma3Config' object has no attribute 'vocab_size' 3 messages
Running dynamic 4-bit models with VLLM 39 messages
Effects of multiple GPUs on training of LLMs 5 messages
Gemma support with pytorch 2.4.0 4 messages
AttributeError: 'Gemma3Processor' object has no attribute 'tokenize' 6 messages
Open
Gemma 3 Finetuning Issues 3 messages
Solved
Unsloth Qwen2.5-1.5B Finetuning Issue 2 messages
Open
Getting errors after Tokenization with Gemma 3 only 7 messages
Looking for tips 49 messages
Solved
`vllm serve` an unsloth bnb Dynamic 4-bit model with tool support 3 messages
Searching for Good Audio Tokenizer 19 messages
SmolVLM2: Unexpected Indentation Error 47 messages
Open
Gemma 3 Colab Notebook not training 6 messages
KeyError: '' During SFTTrainer Creation When Fine Tuning Llama-3.2-11B-Vision-Instruct 4 messages
Solved
vLLM Server Error 3 messages
Unsloth support for Transfer Learning 3 messages
Trying to run Unsloth models using vLLM with Kubernetes. 2 messages
How to enable embed_tokens training in Gemma 3 with Unsloth? 3 messages
Facing Error during inference: Cache only has 0 layers, attempted to access layer with index 0 2 messages
Test Bug fixes without gpus 4 messages
Looking for Advice: Building a Water Management Chatbot with a Fine-Tuned LLM 11 messages
Open
_unsloth_get_batch_samples error when running demo locally 5 messages
Open
issue in inferencing output 3 messages
Open
Should it be this slow? 2 messages
Solved
Failed to load quantized Gemma model from local directory after saving with save_pretrained 4 messages
Open
got an unexpected keyword argument 'num_items_in_batch' 2 messages
Open
Issue on with Docker image as Unsloth not using GPU drivers 5 messages
Gemma3 1b: expected mat1 and mat2 to have the same dtype, but got: float != struct c10::BFloat16 5 messages
New Mistral 3.1 24b model finetune 10 messages
gemma3 loses multimodal support in Ollama when finetuned with Unsloth 50 messages
GRPO training when you have reasoning tokens 17 messages
PeftConfig error: Can't find 'adapter_config.json' at 'unsloth/gemma-3-1b-it-unsloth-bnb-4bit' 3 messages
Last update causing gemma oom 6 messages
Unable to Save Model; mmap can only be used with files 2 messages
Solved
cant save my model in pytorch format therefore cant load it 29 messages
Continued Pre-Training is broken 7 messages
Open
My nightly job 20250319 failed with error VLLM_USE_V1=1 is not supported with --quantization 2 messages
Fast Model Loads 4-bit Variant Even When load_in_4_bit=False 2 messages