Chat with your own models, right in the side panel.
EZGPT.APP puts a chat window for your own language models in the browser side panel. Point it at LM Studio or Ollama on this machine, a server on your LAN, or a cloud API, and talk to the model while you keep browsing. It stays open when you click around the page, and it never sends anything anywhere except the server you configured.
Free. No account. Chrome and Edge 116 or newer.
On localhost or your LAN. The default preset.
LocalOn localhost or your LAN. No extra CORS setup needed.
LocalThrough the chat completions API with your own key.
CloudHundreds of hosted models behind one key.
CloudClaude models through the OpenAI-compatible endpoint.
CloudGoogle models through the OpenAI-compatible endpoint.
CloudvLLM, llama.cpp, LocalAI, text-generation-webui and others.
CustomOpen it with one click or Ctrl+Shift+E. It stays open while you browse, and the reply keeps streaming.
Discover the models a server offers and change model per chat without editing settings.
Partial replies are kept if you stop or close the panel.
DeepSeek, Qwen, QwQ and gpt-oss style thinking shows in a collapsible block, whether the server sends it as a reasoning field or as think tags.
Attach the selected text or the readable text of the current tab to your message. Right-click any selection and choose "Ask EZGPT.APP about…".
Paste, drop or attach screenshots for vision models such as Qwen-VL, LLaVA, Pixtral or GPT-4o.
Code blocks with one-click copy.
Temperature, top-p, max tokens and stop sequences, set for each chat.
Save the prompts you reuse from the composer's tool row and insert them with a click.
Search, rename and grouping by day. Export a chat as Markdown or JSON, or back up everything as one file and restore it later.
Light, dark and system themes, four accents, three densities.
Click "Open in tab" in the top bar to get the same chat in a full tab, with the chat list as a permanent left column.
One click from the official store. Works in Chrome and Edge 116 or newer. There is nothing to sign up for and nothing to configure on a server.
Free • No registration required • Version 4.0
Start your model server, click the EZGPT.APP icon, pick a model in the top bar and type your first message. Pick your server below for the details.
http://localhost:1234/v1 already selected.Running LM Studio on another machine on your LAN? Use its IP instead of localhost and tick "Serve on local network" in LM Studio.
ollama pull llama3.2. Ollama's server is already running once it is installed.http://localhost:11434/v1, then Save provider.OLLAMA_ORIGINS.Your key is stored only in this browser and is sent only to the OpenAI server, as an Authorization header.
These cloud services are reached through their OpenAI-compatible endpoints. Each has a preset, so you only add a key.
Use Alt+1…9 to jump between the providers you have set up.
EZGPT.APP works with any server that speaks the OpenAI chat completions API: vLLM, llama.cpp, LocalAI, text-generation-webui and others, on your machine, your LAN or a host you rent.
/v1, and a key if the server needs one.There is no server on our side. The extension contains no analytics library, no crash reporter, no advertising and no remote code.
Settings, chats and images live in your browser's storage. Uninstalling the extension deletes them.
Requests go only to the servers you add, and only when you send a message, discover models or test a connection.
Page text is read only when you click "Attach page" or use the context menu. Never in the background.
Your providers, chats, images and snippets are migrated automatically the first time 4.0 starts. The popup and detached window are replaced by the side panel. Open it with the toolbar icon or Ctrl+Shift+E. The first time you talk to each server, the browser asks you to allow access to that address. 3.x asked for every website at install; 4.0 asks for nothing until you use something.
Add EZGPT.APP to your browser and chat with your own models from the side panel.
Free • No registration required • Chrome and Edge 116+