Soup CLI Traffic & Growth
Fine-tune an 8B LLM on a 4 GB laptop GPU
Estimated website visits are unavailable
Canonical domain not yet confirmed
About Soup CLI
LoRA keeps the base model frozen: read, never written. So Soup keeps it in system RAM and streams it into the GPU one decoder layer at a time. Peak VRAM becomes one layer instead of the whole model. Measured on an RTX 3050 Laptop 4 GB: Llama-3.1-8B trains at 119.6 tok/s in 3.32 GB peak. One YAML, one command. SFT, DPO, GRPO, KTO, plus eval, gating and export. Apache-2.0. Every number is published, including the ones I measured and threw away.
Frequently Asked Questions
What is Soup CLI?
Soup CLI is LoRA keeps the base model frozen: read, never written. So Soup keeps it in system RAM and streams it into the GPU one decoder layer at a time. Peak VRAM becomes one layer instead of the whole model. Measured on an RTX 3050 Laptop 4 GB: Llama-3.1-8B trains at 119.6 tok/s in 3.32 GB peak. One YAML, one command. SFT, DPO, GRPO, KTO, plus eval, gating and export. Apache-2.0. Every number is published, including the ones I measured and threw away.
Is Soup CLI free?
Yes, Soup CLI is free to use.
What related tools can I explore with Soup CLI?
These tools share a topic with Soup CLI: Google Whisk 2.0, TestSprite 1.0, Ash , Capalyze. Open each profile to review its available information and official site.
