forum
Error loading the saved model
Open
Finetune InternVL2_5-4B
What's the best unrestrained model?
Solved
Nightly version: Unexpected indentation error
Open
LLama 3.1 8B fine-tuning output repeats input
Open
Llama 3B Fine Tuning Notebook Broken Outside Colab
Open
Hi everyone,I am trying to fine-tune the Llama 3.1 8B 4bit model using Unsloth for a specific task
What's the cheapest/best way to host a model?
Puzzle Clarification
Open
Fine-tuning LLM model for Sentiment Analysis
Save model to VLLM
Solved
Invalid device ordinal
Open
Fine tuning, RAG or combination of both?
Trying to# Save to q4_k_m GGUFmodel.push_to_hub_gguf(FINETUNED_MODEL_NAME, tokenizer, quantization
Converting the 16-bit vision export to 4-bit gguf?
Unsuccessful fine-tuning llama 3 and llama 3.1
How to host the model in google colab
Overall Knowledge Questions
Open
Saving only lora adapters as GGUF
push_to_hub_gguf broken?
Install torch, xformers, trl, unsloth in Linux machine venv
Saving files to gguf
Running llms on mixture of cpu and gpus to get more memory
Runtime Error
Fine Tuning LLM without success
SFTTrainer
Multi-class classification using Unsloth
Open
How to Create Own bnb-4bit Models
Help with Training Data for Llama 3.1
OOM Recoverability?
Open
Unsloth v Instructlab
Can't proceed past downloading model
Open
Help with gpu isses
Open
Custom Base Model Suppport for GRPO
SFT IS BROKEN (USING YOUR TUTOR COLAB)
error loading model: error loading model vocabulary: cannot find tokenizer merges in model file
The issue is the embed_tokens & lm_head not trainable, which will cause NaNs.
After i fine tuned a model on unsloth, can i use it on Ollama and Open WebUI?
GRPO not working at all after training?
Vision finetune (qwen2-VL) notebook OOM
Getting tps below expectations (IQ1_S 131GB)
Open
can I directly run these safetensors on ollama?
Porting the colab notebook to paperspace
Open
Getting error "does not appear to have a file named pytorch_model.bin, model.safetensors, tf_model"
Open
Curious OOM when using "unsloth" gradient checkpointing
Open
GRPO Notebook not working on a local 3090 with 4-bit loading enabled.
Solved
CUDA OOM When trying to load Qwen2.5-7B instruct in GRPO notebook
Solved
How to maximize speed with 200GB VRAM, 251GB RAM DDR5 and 56 cores (112 threads) for Q2_K_XL R1?
My model doesn't upload to HF from the notebook
Open
Need a road map on how to implement a search/suggestion based AI