Free
Free
The experience is free to use in supported browsers once the Vicuna model finishes downloading and caching locally.
- Access
- The experience is free to use in supported browsers once the Vicuna model finishes downloading and caching locally.
Research · Vendor site 2026-10-04
Chat Matt Rickard hosts AIM-style browser chat powered by a Vicuna-13B model running locally via WebGPU and Apache TVM without sending conversation data to a server. Conversations persist in local storage after the model downloads to the client.
Developers experimenting with on-device LLM chat who run desktop Google Chrome 113+ with WebGPU enabled.
WebGPU support is limited to desktop Chrome 113+, initial model download may take minutes, and conversations stay on-device rather than syncing across browsers.
This demo pairs nostalgic AIM styling with modern WebGPU inference so researchers can experiment with fully client-side LLM chat and local conversation caching.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| On-device inference | About copy states the large language model runs entirely in the browser with WebGPU and no server-side conversation data. | Facts sourced | 2026-10-04 |
| Model | Interface defaults to Vicuna-13B trained by LMSys with Apache TVM and MLC Relax runtime components. | Facts sourced | 2026-10-04 |
| Persona | Preset system prompt simulates a 2000s AIM chatbot persona named SmartestChild with lowercase, emoji-heavy replies. | Facts sourced | 2026-10-04 |
Understand the total cost
Free
Free
The experience is free to use in supported browsers once the Vicuna model finishes downloading and caching locally.
GetOData offers an AI-powered Chrome extension for extracting structured data from websites using custom data points, pagination, and multi-format exports. It complements GetOData’s broader API directory and Apify actors for larger scraping workflows.
Explore toolChatGPT Exporter is a Chrome extension for saving ChatGPT conversations as PDF, Markdown, text, CSV, JSON, or images. It supports partial exports, customizable filenames and PDF themes, and preserves advanced ChatGPT outputs such as code blocks, tables, canvas, and math formatting.
Explore toolaitext.chat lets teams build custom GPT-style assistants trained on their own documents, websites, and files, then embed chatbots or iframes on unlimited sites. Plans scale assistants, datasets, storage, monthly credits, analytics, and optional white-label branding.
Explore toolThe practical questions
Chat Matt Rickard hosts AIM-style browser chat powered by a Vicuna-13B model running locally via WebGPU and Apache TVM without sending conversation data to a server. Conversations persist in local storage after the model downloads to the client.
The experience is free to use in supported browsers once the Vicuna model finishes downloading and caching locally.. This record lists ongoing free access; check the plan limits before starting.
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Not confirmed. API access and subscription access may have different terms; consult the linked sources.
WebGPU support is limited to desktop Chrome 113+, initial model download may take minutes, and conversations stay on-device rather than syncing across browsers.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.