The best AI models, all in one place.
EchoLLM is a privacy-focused desktop client for chatting with multiple AI providers. Your API keys stay on your device — nothing is sent to third-party servers.
| Model | Provider | Context | Input | Output |
|---|---|---|---|---|
| Gemini 3 Pro | Google / DeepMind | 1M | $2/M | $12/M |
| Gemini 3.0 Flash | Google / DeepMind | 1M | $0.50/M | $4/M |
| Gemini 2.5 Pro | Google / DeepMind | 1M | $1.20/M | $8/M |
| Gemini 2.5 Flash | Google / DeepMind | 1M | $0.40/M | $3/M |
| GPT-5.2 | OpenAI | 400K | $5/M | $25/M |
| GPT 4.1 | OpenAI | 128K | $3/M | $18/M |
| GPT 4o | OpenAI | 128K | $2/M | $15/M |
| Claude Opus 4.6 | Anthropic | 1M (beta) | $5/M | $25/M |
| Claude 4.5 Sonnet | Anthropic | 200K | $3/M | $15/M |
| Claude 4.0 Sonnet | Anthropic | 400K | $4/M | $20/M |
| Grok 3 | xAI | 512K | $0.80/M | $0.30/M |
| Grok 3 mini | xAI | 64K | $0.20/M | $1.50/M |
| Grok 4 Fast | xAI | 2M | $0.50/M | $3/M |
- Multi-model chat — Switch between 13 models from 4 providers in a single conversation
- Local key management — API keys stored securely on-device via GetStorage
- Persistent chat history — Conversations saved locally with Hive
- Markdown rendering — Full markdown support with syntax highlighting in responses
- Configurable font size — Adjust text scale from 80% to 150%
- Send on Enter — Toggle Enter-to-send behavior
- Model browser — View model details including context window, pricing, and capabilities
- Flutter SDK (3.6+)
- API key from at least one supported provider
git clone https://github.com/thatlinuxguyyouknow/EchoLLM.git
cd EchoLLM
flutter pub get
flutter run -d linux # or macos, windowssnap install echollm| State Management | Provider |
| Local Storage | Hive (chat history), GetStorage (preferences & keys) |
| Markdown | flutter_markdown |
| Fonts | Google Fonts (Ubuntu) |
| HTTP | http package |
Apache-2.0