Fine-tune Qwen3.8-27B for free using Kaggle GPUs and Unsloth
danielhanchen · x · 2026-08-25
You can fine-tune the Qwen3.8-27B model completely for free using a Google account on Kaggle, which offers 30 hours of free GPU (2× Tesla T4). By using Unsloth's QLoRA kernels, the 27B model fits in 24GB VRAM with no accuracy loss, offering 1.5x faster training and 50% less memory usage.
More from coding & agent
- Two types of AI engineers: Ship fast with bugs vs. build slowly with maintainability — NachoSoto · 2026-08-25
- User opinion: Cursor is all you need for coding — tlakomy · 2026-08-25
- Perplexity's Local Computer Harness Tops BrowseComp Benchmarks — AravSrinivas · 2026-08-25
- Agent traces cited as the most valuable company context — nptacek · 2026-08-25
- Perplexity demonstrates fully local agent runtime on DGX Spark — AravSrinivas · 2026-08-25
- Perplexity launches local Portable Computer with PPLX 27B — testingcatalog · 2026-08-25