Ollama Client
Ollama Client – Chat with Local LLMs Inside Your Browser A lightweight, privacy‑first Chrome extension to chat with local LLMs via Ollama, LM Studio, and llama.cpp. Supports streaming, stop/regenerate, RAG, and easy model switching — all without cloud APIs or data leaks.
Last verified:
What is Ollama Client?
Ollama Client is a privacy-first Chrome extension that enables users to chat with locally hosted LLMs directly in their browser. It connects to self-hosted LLM backends including Ollama, LM Studio, and llama.cpp servers, providing a browser-based frontend UI for local LLM conversations without cloud inference. The extension supports multi-provider workflows, allowing users to route models by provider while maintaining a single chat experience.
Key features include provider and model management (connect multiple local servers, switch models, view provider status), chat and session management with streaming responses, stop/regenerate functionality, and session history saved locally via IndexedDB. It supports file attachments (PDFs, text, DOCX) for local RAG workflows with semantic search, advanced content extraction with lazy loading for web pages, and built-in text-to-speech using the Web Speech API. The extension offers prompt templates, customizable model parameters (temperature, top_k, top_p, repeat penalty), and export/import chat sessions as PDF or JSON.
Ollama Client is designed for developers working with local AI models, researchers evaluating self-hosted LLMs, students learning with offline AI chat, and privacy-conscious users who want to avoid cloud services. With 1,000+ users and a 4.8-star rating on the Chrome Web Store, it provides a lightweight, open-source solution for private AI conversations. The extension is free and open source under the MIT license, with all data stored locally and no external data transfer by default.
Ollama Client pricing
Pricing model: Freemium
Free and open source under the MIT license. Available as a free Chrome extension from the Chrome Web Store (version 0.6.0, 2.51MiB size). No paid tiers, subscriptions, or premium plans. The extension does not collect or sell user data, and all inference runs locally on the user's machine. Optional remote providers like OpenAI require users to bring their own API keys.
Ollama Client pros
- 100% local-first with no cloud inference by default
- Supports Ollama, LM Studio, and llama.cpp providers
- No API keys required for local Ollama connection
- Streaming responses with stop and regenerate buttons
- Multi-chat sessions saved locally via IndexedDB
- PDF, text, and DOCX file uploads for local RAG
- Built-in text-to-speech with voice selection and pitch/rate controls
- Export chat sessions as PDF or JSON
- Import chat sessions from JSON files
- Model search and pull directly in the UI with progress indicator
- Model deletion with confirmation to manage storage
- Load/unload models to manage Ollama memory footprint
- Custom prompt templates and model parameter tuning
- Advanced content extraction with lazy loading for web pages
- Open source under MIT license on GitHub
- Works on Chrome, Brave, Edge, Opera, and Chromium browsers
- Automatic CORS handling via Declarative Net Request
- Responsive UI with Shadcn UI design
- Session metadata includes titles and timestamps
- Test connection button in provider settings
Ollama Client cons
- Requires local Ollama/LM Studio/llama.cpp server to be running
- Performance depends on user's hardware and backend server
- Extension does not include models or run inference itself
- Stop Generation doesn't always abort promptly
- Stop Pull during model downloads may glitch
- CORS may fail on older Chromium/Firefox versions
- Large chat histories in IndexedDB can impact performance
- Firefox support is experimental and requires manual permissions setup
Frequently asked questions about Ollama Client
What is Ollama Client?
Ollama Client is a privacy-first Chrome extension that lets you chat with locally hosted LLMs using Ollama, LM Studio, or llama.cpp — no cloud, no tracking. It's a browser-based frontend UI for local LLM servers that connects to your self-hosted LLM backend and lets you chat inside your browser. All data stays on your machine and local network by default.
Do I need an API key to use Ollama Client?
No API key is required for local Ollama, LM Studio, or llama.cpp connections. The extension connects to your local server at localhost:11434 (Ollama), localhost:1234/v1 (LM Studio), or localhost:8000/v1 (llama.cpp). Only remote providers like OpenAI require you to bring your own API key.
Is my data private when using Ollama Client?
Yes. Ollama Client is local-first by design with no cloud inference, no external data transfer, and no first-party analytics or telemetry. Chat prompts, responses, session metadata, and settings are stored locally in browser storage backed by SQL WASM. Your data only leaves your device if you configure a remote provider endpoint.
Which browsers support Ollama Client?
Ollama Client works with any Chromium-based browser including Chrome, Brave, Edge, Opera, and Chromium. Firefox support is available via temporary addon installation but requires manual permissions setup and is considered experimental.
How do I set up Ollama Client?
Setup is simple: 1) Install the extension from the Chrome Web Store, 2) Install and run Ollama from ollama.com, 3) Pull at least one chat model and one embeddings model for RAG, 4) Open the extension's options page, go to the Providers tab, enable your provider, set the base URL, and test the connection, 5) Open the side panel, select a model, and start chatting.
What file types can I upload for RAG?
The extension supports PDFs, text files, and DOCX files. Uploaded files are automatically chunked and embedded for semantic search and local RAG workflows. Plain text-based formats like .txt, .md, .js, and .ts are also supported for direct reading.
Can I export my chat history?
Yes. You can export single or all chat sessions as PDF or JSON files. Chat history is also importable from JSON files. Session data is saved locally using IndexedDB, and you can export optional ZIP backups stored locally.
What models work best with Ollama Client?
Recommended models depend on your hardware: 8GB RAM (no GPU) works with gemma:2b or mistral:7b-q4; 16GB RAM works with gemma:3b-q4 or mistral; 16GB+ with 6GB VRAM works with llama3:8b-q4 or gemma:3b; RTX 3090+ or Apple M3 Max can run llama3:70b or mixtral. Quantized models (q4_0, q5_1) are preferred for better performance.
How do I handle CORS issues?
Ollama Client uses Declarative Net Request (DNR) for automatic CORS handling in Chrome-based browsers. If you encounter CORS issues, follow the setup guide at ollama-client.shishirchaurasiya.in/ollama-setup-guide. CORS may still fail on older Chromium/Firefox versions or under certain network policies.
Can I use Ollama Client without internet?
Yes, once set up, Ollama Client works offline for local inference. The extension connects to your local LLM server running on your machine, and all data processing happens locally. You only need internet to initially download the extension, install Ollama, and pull models from the Ollama registry.