forum
Fine-tune video in Qwen3
Issue: GGUF conversion crashes Colab system
Lora Parameter r
Inference Endpoints Unsloth Models
Open
Export my Qwen3 VL 8B Instruct adapter (after merging) into GGUF
Hi, I am getting an error when trying to save my finetuned Gemma3 27B model
issue in trainer
[solved] Qwen3 VL Index out of bounds
Solved
[solved] Qwen3 VL Index out of bounds
Solved
llama.cpp slow
Multi-GPU gpu utilization only uses one gpu at a time
Qwen3 Vl 30b
Infinite Waiting for Unsloth Installation
Trouble running Advanced_Llama3_2_(3B)_GRPO_LoRA.ipynb via Apptainer on HPC
KeyError:'align_logprobs_with_mask
data preprocess in STT
Unsloth: Will smartly offload gradients to save VRAM!
i suddenly ran into this error in the STT notebook
tips for getting better training
Adapters format when finetuned using QLoRA
Qwen3-VL-32B?
gpt-oss-20b export model for vLLM
Open
Solved
plan on supporting deepseek-ocr on all the tuning?
Qwen 2.5 VL always takes up all the vram on all gpus
Open
How to set Video FPS for Inference and Finetuning
how to set image max pixels
is LLM learning?
FailOnRecompileLimitHit on GptOssForCausalLM_forward
I'm looking for some input from someone that is good at finetuning
Not sure if lora gets used for inference
gpt-oss-(20B)-GRPO Triton Compiler Error
Solved
VLLM VRAM Usage GRPO
Example inference code for Qwen3-VL 8B Instruct model (Vision + Language) using Unsloth
Gemma3-1b-it fails with vllm + fullft + fast_inference
how to use the finetuned model.
Why ??
Open
NameError: name 'slice_indices' is not defined.
Wondering how to include printing the validaltion loss during finetuning
Inquiry About Training a Language Model on a Local Arabic Dialect
Does unsloth have a provision to support diffusion transformers?
SFT model giving wrong answers even after full reset (Llama-3-8B 4-bit, Unsloth SFTTrainer)
Ling-1t - Why is Q5 smaller than Q8?
QAT training query
Open
NotImplementedError: Cannot copy out of meta tensor; no data!
Open
python segmentation fault
How can I sft GPT OSS with limited hardware
General PEFT support
Solved
Clarification on best dataset format for Unsloth fine-tuning
Windows + AMD setup?
Support InternVL3 patch
Open