{"data":{"slug":"kochava-studios-witsy","name":"witsy","tagline":"Desktop AI Assistant and Universal MCP Client","github_url":"https://github.com/Kochava-Studios/witsy","owner":"Kochava-Studios","repo":"witsy","owner_avatar_url":"https://avatars.githubusercontent.com/u/258930584?v=4","primary_language":"TypeScript","stars":2026,"forks":171,"topics":["anthropic","deepseek","electron-app","electronjs","gemini","genai","googleai","groq","mcp","mcp-client","mistral","mistralai","ollama","ollama-gui","openai","openrouter","vuejs","vuejs3"],"archived":false,"github_pushed_at":"2026-04-23T17:32:21+00:00","maintenance_label":"Slowing","stars_delta_30d":10,"url":"https://www.graphcanon.com/tools/kochava-studios-witsy","markdown_url":"https://www.graphcanon.com/tools/kochava-studios-witsy.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/kochava-studios-witsy","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=kochava-studios-witsy","description":"Witsy: desktop AI assistant / universal MCP client","homepage_url":null,"license":"AGPL-3.0","open_issues":60,"watchers":15,"ai_summary":"A desktop application that acts as an AI assistant and can run models from various LLM providers through MCP servers.","readme_excerpt":"
\n\n \"Witsy\n
Witsy
\n
Desktop AI Assistant
Universal MCP Client
\n\n
\n\n

\n
\n\n\n\n\n\n\n
\n\n## Downloads\n\nDownload Witsy from the [releases](https://github.com/Kochava-Studios/witsy/releases) page.\n\nOn macOS you can also `brew install --cask witsy`.\n\n\n## What is Witsy?\n\nWitsy is a BYOK (Bring Your Own Keys) AI application: it means you need to have API keys for the LLM providers you want to use. Alternatively,\nyou can use [Ollama](https://ollama.com) to run models locally on your machine for free and use them in Witsy.\n\nIt is the first of very few (only?) universal MCP clients:
***Witsy allows you to run MCP servers with virtually any LLM!***\n\n## Supported AI Providers\n\n| Capability | Providers |\n|------------|-----------|\n| **Chat** | OpenAI, Anthropic, Google (Gemini), xAI (Grok), Meta (Llama), Ollama, LM Studio, MistralAI, DeepSeek, OpenRouter, Groq, Cerebras, Azure OpenAI, any provider who supports the OpenAI API standard (together.ai for instance)|\n| **Image Creation** | OpenAI, Google, xAI, Replicate, fal.ai, HuggingFace, Stable Diffusion WebUI |\n| **Video Creation** | OpenAI, Google, Replicate, fal.ai |\n| **Text-to-Speech** | OpenAI, ElevenLabs, Groq, fal.ai |\n| **Speech-to-Text** | OpenAI (Whisper), fal.ai, Fireworks.ai, Gladia, Groq, nVidia, Speechmatics, Local Whisper, Soniox (realtime and async) any provider who supports the OpenAI API standard |\n| **Search Engines** | Perplexity, Tavily, Brave, Exa, Local Google Search |\n| **MCP Repositories** | Smithery.ai\n| **Embeddings** | OpenAI, Ollama |\n\nNon-exhaustive feature list:\n- Chat completion with vision models support (describe an image)\n- Text-to-image and text-to video\n- Image-to-image (image editing) and image-to-video\n- LLM plugins to augment LLM: execute python code, search the Internet...\n- Anthropic MCP server support\n- Scratchpad to interactively create the best content with any model!\n- Prompt anywhere allows to generate content directly in any application\n- AI commands runnable on highlighted text in almost any application\n- Experts prompts to specialize your bot on a specific topic\n- Long-term memory plugin to increase relevance of LLM answers\n- Read aloud of assistant messages\n- Read aloud of any text in other applications\n- Chat with your local files and documents (RAG)\n- Transcription/Dictation (Speech-to-Text)\n- Realtime Chat aka Voice Mode\n- Anthropic Computer Use support\n- Local history of conversations (with automatic titles)\n- Formatting and copy to clipboard of generated code\n- Conversation PDF export\n- Image copy and download\n\n

\n   \n   \n \n

\n\n## Setup\n\nYou can download a binary from from the [releases](https://github.com/Kochava-Studios/witsy/releases) page or build yourself:\n\n```\nnpm ci\nnpm start\n```\n\n## Prerequisites\n\nTo use OpenAI, Anthropic, Google or Mistral AI models, you need to enter your API key:\n- [OpenAI](https://platform.openai.com/api-keys)\n- [Anthropic](https://console.anthropic.com/settings/keys)\n- [Google](https://aistudio.google.com/app/apikey)\n- [xAI](https://console.x.ai/team/)\n- [Meta](https://llama.developer.meta.com/api-keys/)\n- [MistralAI](https://console.mistral.ai/api-keys/)\n- [DeepSeek](https://platform.deepseek.com/api_keys)\n- [OpenRouter](https://openrouter.ai/settings/keys)\n- [Groq](https://console.groq.com/keys)\n- [Cerebras](https://cloud.cerebras.ai/platform/)\n\nTo use Ollama models, you need to install [Ollama](https://ollama.com) and download some [models](https://ollama.com/search).\n\nTo use text-to-speech, you need an \n- [OpenAI API key](https://platform.openai.com/api-keys).\n- [Fal.ai API Key](https://fal.ai/dashboard/keys)\n- [Fireworks.ai API Key](https://app.fireworks.ai/settings/users/api-keys)\n- [Groq API Key](https://co","github_created_at":"2024-04-25T03:55:22+00:00","created_at":"2026-07-15T11:18:27.366888+00:00","updated_at":"2026-09-20T05:17:34.358012+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"}],"tags":[{"slug":"anthropic","name":"anthropic"},{"slug":"deepseek","name":"deepseek"},{"slug":"electronjs","name":"electronjs"},{"slug":"gemini","name":"gemini"},{"slug":"genai","name":"genai"},{"slug":"googleai","name":"googleai"},{"slug":"groq","name":"groq"},{"slug":"mcp-client","name":"mcp-client"}],"trust":{"provenance":{"is_fork":false,"github_id":791622368,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T05:17:32.083Z","maintenance":{"label":"Slowing","score":36,"methodology":"github_public_v1","releases_90d":0,"days_since_push":149,"last_release_at":"2026-03-04T23:37:59Z","stars_delta_30d":10,"open_issues_delta_30d":5},"security_summary":{"status":"no_manifest","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-15T11:18:28.773Z","medium_count":0,"scan_profile":"mcp_manifest","critical_count":0}},"capability_facts":{"mcp":{"source":"package.json:@modelcontextprotocol/*","observed_at":"2026-09-20T05:17:33.396Z","server_manifest":true},"scan":{"source":"repo_scan","observed_at":"2026-09-20T05:17:33.396Z"},"has_cli":{"value":true,"source":"package.json:bin|scripts","observed_at":"2026-09-20T05:17:33.396Z"},"languages":{"value":["typescript","javascript"],"source":"github.language+package.json","observed_at":"2026-09-20T05:17:33.396Z"},"license_spdx":{"value":"AGPL-3.0","source":"github.license","observed_at":"2026-09-20T05:17:33.396Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["If you need an AI solution that can integrate with almost any LLM provider via MCP servers.","When your workflow heavily relies on a wide array of services from different providers such as OpenAI for chat completion and HuggingFace for image creation.","For users who require the ability to bring their own keys (BYOK) for accessing proprietary or paid models while also having the option to run free, local models through Ollama."],"when_not_to_use":["If you do not have existing API keys from LLM providers and are not interested in setting up a local model through Ollama.","For users who require an out-of-the-box solution that does not necessitate managing multiple provider integrations or configurations for different services."],"source":"enrich:decision_facts","observed_at":"2026-07-17T12:46:39.762Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Witsy is a desktop AI assistant and universal MCP client that supports numerous LLM providers. It requires users to have API keys for models they wish to use, but also supports local model running through Ollama."}]}}