{"data":{"slug":"fikrikarim-parlor","name":"parlor","tagline":"On-device real-time multimodal AI for voice and vision","github_url":"https://github.com/fikrikarim/parlor","owner":"fikrikarim","repo":"parlor","owner_avatar_url":"https://avatars.githubusercontent.com/u/13066728?v=4","primary_language":"Python","stars":2068,"forks":272,"topics":["apple-silicon","gemma","kokoro","litert-lm","local-llm","mlx","multimodal","on-device-ai","python","real-time","speech-recognition","text-to-speech","voice-assistant"],"archived":false,"github_pushed_at":"2026-08-03T18:06:56+00:00","maintenance_label":"Steady","stars_delta_30d":154,"url":"https://www.graphcanon.com/tools/fikrikarim-parlor","markdown_url":"https://www.graphcanon.com/tools/fikrikarim-parlor.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/fikrikarim-parlor","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=fikrikarim-parlor","description":"On-device, real-time multimodal AI with features similar to GPT-Live","homepage_url":null,"license":"Apache-2.0","open_issues":1,"watchers":21,"ai_summary":"Parlor enables local running of natural voice and vision conversations using Gemma 4 E2B and Kokoro models, focusing on real-time interactivity with minimal external dependencies.","readme_excerpt":"## Requirements\n\n- Python 3.12+\n- [llama.cpp](https://github.com/ggml-org/llama.cpp) (`brew install llama.cpp` on macOS; other platforms: [install guide](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)). Needs build b9503 (June 2026) or newer, b9512 for `MODEL=12b` — older builds lack Gemma 4 audio or crash loading its mmproj\n- macOS with Apple Silicon, or Linux with a supported GPU\n- ~6 GB free RAM for the default E4B model (`MODEL=e2b` fits in ~4 GB)\n\n---\n\n## Quick start\n\n```bash\ngit clone https://github.com/fikrikarim/parlor.git\ncd parlor\n\n---\n\n# Install uv and llama.cpp if you don't have them\ncurl -LsSf https://astral.sh/uv/install.sh | sh\nbrew install llama.cpp\n\nuv sync\nuv run parlor\n```\n\nOpen [http://localhost:8000](http://localhost:8000), grant camera and microphone access, and start talking.\n\nModels are downloaded automatically on first run (~5.7 GB for Gemma 4 E4B QAT + its multimodal projector, plus TTS models).","github_created_at":"2026-04-05T17:33:55+00:00","created_at":"2026-07-11T12:09:32.067205+00:00","updated_at":"2026-09-20T07:09:54.82449+00:00","categories":[{"slug":"computer-vision","name":"Computer Vision","url":"https://www.graphcanon.com/categories/computer-vision","markdown_url":"https://www.graphcanon.com/categories/computer-vision.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/computer-vision"},{"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"},{"slug":"speech-audio","name":"Speech & Audio","url":"https://www.graphcanon.com/categories/speech-audio","markdown_url":"https://www.graphcanon.com/categories/speech-audio.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/speech-audio"}],"tags":[{"slug":"apple-silicon","name":"apple-silicon"},{"slug":"gemma","name":"gemma"},{"slug":"kokoro","name":"kokoro"},{"slug":"litert-lm","name":"litert-lm"},{"slug":"local-llm","name":"local-llm"},{"slug":"mlx","name":"mlx"},{"slug":"multimodal","name":"multimodal"},{"slug":"on-device-ai","name":"on-device-ai"}],"trust":{"provenance":{"is_fork":false,"github_id":1202171123,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-20T07:09:52.032Z","maintenance":{"label":"Steady","score":60,"methodology":"github_public_v1","releases_90d":2,"days_since_push":47,"last_release_at":"2026-08-02T19:20:18Z","stars_delta_30d":154,"open_issues_delta_30d":-8},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T12:09:34.451Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-20T07:09:53.456Z"},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-09-20T07:09:53.456Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-09-20T07:09:53.456Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-20T07:09:53.456Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["Need to run complex, interactive AI locally without relying on cloud services.","Specific hardware requirements like Apple Silicon or Linux with supported GPUs do not pose a barrier."],"when_not_to_use":["Limited by Python 3.12 requirement and need for specialized hardware such as Apple Silicon.","Insufficient ~3 GB RAM makes it unsuitable for environments with tight memory constraints."],"source":"enrich:decision_facts","observed_at":"2026-07-17T12:23:29.163Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Parlor is an all-in-one multimodal AI solution running entirely on device for real-time voice and vision interactions."}]}}