February 20th, 2026
QuickWhisper 1.25
We're thrilled to announce QuickWhisper 1.25 - a release built for power users, developers, and anyone who wants deeper control over their transcription workflow. This update brings custom model support, a local API server, and seamless integrations with AI agents.
New Features
Add Custom Whisper Models (GGML format)
QuickWhisper now supports importing custom Whisper models in GGML format. Download any fine-tuned model from popular hubs like HuggingFace and load it directly into QuickWhisper. This long-awaited feature is a game-changer for users working with specialized vocabularies, niche languages, or domain-specific content where standard Whisper models fall short.
Developer Mode & Local REST API
This is one we're extremely excited about. Spin up Local API Server inside QuickWhisper. This gives third-party tools and scripts programmatic access to your transcripts and summaries. Even better, you can copy a ready-to-use mpc.json configuration and paste it into your AI agent settings, turning QuickWhisper into a local tool for your AI workflows.
Deep Link Support
Open any specific transcript directly from an external link - no need to navigate through the UI. This is especially powerful when combined with the local API: query transcript IDs programmatically, then open them instantly via deep links for a seamless automated workflow.
Example:
quickwhisper://open?id=AF760979-EB59-49E2-891B-ECAABACDCB43Cloud Configuration Testing
We've added a dedicated test button for both Cloud Voice Models and Cloud Intelligence models. Instantly verify that your API credentials are correct and that a connection to your cloud provider can be established. This dramatically improves the configuration experience and helps you catch errors early.
These features mark a significant step toward making QuickWhisper not just a standalone app, but a core building block in your broader productivity and AI toolchain. As always, we welcome your feedback and suggestions for future improvements!
The QuickWhisper Team