Soup social preview
fine-tuning10,692Apache 2.0

Soup

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

Updated Nov 21, 2026
Platforms
Pricing
free-open-source
Status
active
License
Apache 2.0

What it does

Core capabilities at a glance

  • CLI
  • Consumer GPU
  • DPO
  • Fine Tuning
  • Gguf
  • Huggingface
  • Llmops
  • Local AI

Deep dive

The full breakdown - performance, comparisons, and setup

SoupLink to this section

Soup is a fine-tuning toolkit - Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

OverviewLink to this section

Fine-tune and post-train LLMs in one command. No SSH, no config hell.

Soup turns the pain of LLM fine-tuning into a simple workflow. One config, one command, done.

Fine-tune an 8B model on a 4 GB laptop GPU. Layer streaming keeps the frozen base out of VRAM and feeds it to the GPU one decoder layer at a time. Measured on an RTX 3050 Laptop 4 GB: Llama-3.1-8B-Instruct + NF4 at 119.6 tok/s, 3.32 GB peak — bit-exact against a normal resident run, and reproduced independently on an H100 at 113.00 tok/s in the same 3.32 GB. (The tok/s figure was measured on v0.72.2, before the v0.73.0 correctness repair that cost −4.8% at 32B; it has not been re-run on a 4 GB card since.) Opt-in ('stream_layers: true') and still BETA — how it works · all measurements · paper · check it yourself on a free Colab T4 (caps the process to 4 GB, then asserts a streamed model is bit-identical to a normal one)

Training LLMs is still painful. Even experienced teams spend 30-50% of their time fighting infrastructure instead of improving models. Soup fixes that.

Soup is open-source, written primarily in Python, with 10,692 GitHub stars under the Apache 2.0 license. The latest release is v0.73.2 (2026-08-15).

Key capabilitiesLink to this section

From the project's documentation:

  • Zero SSH. Never SSH into a broken GPU box again.
  • One config. A simple YAML file is all you need.
  • Auto everything. Batch size, GPU detection, quantization — handled.
  • Works locally. Train on your own GPU with QLoRA. No cloud required.
  • A suite scored 0.225 for a model that got it right 40/40. mini_tool_call was
  • New: a benign-prompt axis. Leg 2 flagged a drop in refusal rate and had no reverse,

InstallLink to this section

A quick way to get started (always check the official docs for the latest):

pip install soup-cli

How it fits a local-AI stackLink to this section

Soup runs on your own hardware, so pair it with a model and a GPU sized to your needs. Use the VRAM calculator to pick a model that fits your card, and see what you can run for hardware guidance. Related fine-tuning toolkits in the directory:

SourcesLink to this section

Stats from GitHub, 2026-08-17.

From the community

What users are saying

“Soup up originated in the U.S. in the late 19th century, though it wasn't widely used until the 20th century. Its exact origins are unknown, but it could be short for supercharge, or it might come from a horse-racing slang term for injecting horses with narcotics meant to make them run faster. Interesting.”

“In the soup story the villagers freely gave up their carrots and onions and the travelers didn't give any guarantees that they wouldn't be consumed.”

“I know you joke, but this advert is surely completely untargeted? It's definitely untargeted in the modern meaning (they don't pick which adverts to insert into the book when you buy it based on a profile of you).”

Frequently asked

Quick answers to common questions

What is Soup?

Soup is a fine-tuning tool for local AI workloads. Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

Is Soup free and open source?

Yes, Soup has 10,692 GitHub stars and is licensed under Apache 2.0. You can self-host it for free on .

What hardware do I need for Soup?

The hardware requirements depend on which models you run. Check our hardware directory for compatible GPUs and systems. Soup has 10,692 GitHub stars and an active community.

Does Soup support GPU acceleration?

Soup's GPU support depends on your specific setup. Check the documentation for details. For the best performance, pair it with an NVIDIA RTX 4090 or 5090.

What are the best alternatives to Soup?

Popular alternatives include other fine-tuning tools in our directory. Browse our full collection at /tool for comparisons, community reviews, and benchmark data to find the right fit for your workflow.

How much does Soup cost?

Soup is free-open-source. It is completely free and open source to self-host.

Pairs well with

Complementary tools, models, and hardware