Soup CLI vs AI Eyes
Comprehensive feature analysis, ratings breakdown, platform compatibility, and community review comparison.
Detailed Feature Comparison Matrix
Compare Other Tools| Dimension |  Soup CLI |  AI Eyes |
|---|---|---|
| Primary Category | Open Source | Open Source |
| Community Rating | No ratings(0 reviews) | No ratings(0 reviews) |
| Community Upvotes | 104 votes | 63 votes |
| Supported Platforms | Web | Web |
| Tags & Focus | #Artificial Intelligence#Developer Tools#Open Source#GitHub | #Developer Tools#Open Source#Privacy#OpenAI Day |
| Maker / Company | Independent Developer | Independent Developer |
| Platform Verification | Community Listing | Community Listing |
About Soup CLI
Soup CLI is an innovative, open-source developer tool designed to democratize LLM fine-tuning by enabling developers to train large language models on modest consumer hardware. Specifically, Soup CLI makes it possible to fine-tune an 8B LLM using just a 4 GB laptop GPU.
Because Low-Rank Adaptation (LoRA) keeps the base model frozen—meaning it is read from and never written to—Soup CLI keeps the base model in system RAM and streams it into the GPU one decoder layer at a time. This reduces peak VRAM consumption to just a single layer instead of holding the entire model in memory. Tested on an RTX 3050 Laptop 4 GB GPU, Llama-3.1-8B trains at an impressive 119.6 tokens/sec while maintaining a peak usage of only 3.32 GB.
Featuring a streamlined workflow requiring just one YAML configuration file and a single command, Soup CLI supports multiple training methods including SFT, DPO, GRPO, and KTO, alongside built-in capabilities for evaluation, gating, and model export. Distributed under the Apache-2.0 license, the project prides itself on radical transparency, publishing all performance numbers—including experimental data that didn't make the cut.
Pros of Soup CLI
- Fine-tune an 8B LLM (like Llama-3.1-8B) on a modest 4 GB laptop GPU
- Extremely memory efficient by streaming the frozen base model layer-by-layer from system RAM
- Simple user experience with a single YAML configuration file and one-command execution
- Supports multiple training methods including SFT, DPO, GRPO, and KTO, plus eval, gating, and export
- Open-source (Apache-2.0) with transparent performance data and published metrics
Cons of Soup CLI
- Requires sufficient system RAM to hold the base model while streaming to the GPU
- Performance heavily relies on system memory bandwidth for layer streaming
- Limited to hardware configurations compatible with the CLI's memory-streaming architecture
Frequently Asked Questions
What is Soup CLI?
Soup CLI is an open-source developer tool that allows users to fine-tune large language models (such as an 8B LLM) on a low-end 4 GB laptop GPU by leveraging a layer-by-layer memory streaming approach.
How does Soup CLI reduce VRAM usage?
Because LoRA keeps the base model frozen and read-only, Soup CLI stores the base model in system RAM and streams it into the GPU one decoder layer at a time. This keeps peak VRAM consumption down to just a single layer rather than the whole model.
What training methods and features are supported?
Soup CLI supports SFT (Supervised Fine-Tuning), DPO, GRPO, and KTO. It also includes built-in evaluation, model gating, and model export functionalities, all configured via a single YAML file and executed with one command.
Is Soup CLI open-source?
Yes, Soup CLI is released under the Apache-2.0 open-source license.
About AI Eyes
AI Eyes is a permission-first co-presence prototype designed for AI companions. Created by developer Deen Storkey, this open-source tool establishes a privacy-focused, local-first foundation for managing what your digital companions can see and hear.
With its initial v0.1 release, AI Eyes allows users to choose a specific screen or window to capture and gives granular, separate controls for screen, system audio, and microphone inputs. The interface makes it easy to invite named companions and ensures transparency by always displaying when sensing is active. Users can pause or end sessions at any time; ending a session immediately stops all media tracks and resets the session completely.
Built using Codex and GPT-5.6, this prototype is focused entirely on local-first permission, presence, and session-control. Please note that in this foundational stage, no external AI provider or semantic understanding has been connected yet.
Pros of AI Eyes
- Privacy-focused, local-first permission architecture
- Separate, granular control over screen, system audio, and microphone capture
- Clear visual indicators showing when active sensing is occurring
- Open-source code base allowing developers to review and build upon the prototype
Cons of AI Eyes
- It is a prototype (v0.1) and does not yet have an active AI provider or semantic understanding connected
- Limited to foundational capture and session-control features at this stage
Frequently Asked Questions
What is AI Eyes?
AI Eyes is an open-source, permission-first co-presence prototype built to manage what AI companions can access. It provides local-first controls over screen sharing, audio, and session state.
Does AI Eyes currently analyze my screen or voice?
No. In the v0.1 release, no AI provider or semantic understanding is connected. The prototype is currently a local-first foundation focused on permission, capture, presence, and session-control.
How do I control my privacy while using AI Eyes?
AI Eyes lets you choose exactly which screen or window to capture and gives you separate toggle controls for your screen, system audio, and microphone. There is also a clear indicator to show when sensing is active, and you can pause or end the session to stop all media tracks instantly.
What technologies were used to build AI Eyes?
The AI Eyes prototype was built using Codex and GPT-5.6.