Meetily Pro 1.5.0 - Your Most Requested Features Release Notes

What's New in Pro 1.5.0

This is the biggest Pro feature release since 1.1.1. It ships the two most requested capabilities from user feedback - audio file import and custom summary templates - alongside a complete UI redesign and a new default AI model that runs on half the RAM.

158 files changed, 15 PRs merged, ~126 non-merge commits. 13 days of development.

Import & Retranscribe

Drag-and-drop any audio or video file to transcribe it, or retranscribe a past meeting with a different model. Available in both Community and Pro editions.

10 supported formats - MP4, M4A, WAV, MP3, FLAC, OGG, AAC, MKV, WebM, WMA
Drag-and-drop - Drop a file onto the app to start a new transcription
Retranscribe existing meetings - Recorded with the wrong model? Re-run transcription with a different provider (export your original first via Share)
Parallel processing - Chunked resampling via Rayon for faster import of large files
Smart decoding - Symphonia decoder with FFmpeg fallback for maximum format compatibility

This was F-6 in our prioritization doc - ranked HIGH as a recovery mechanism for users who experienced transcript issues. It also unlocks the use case of transcribing pre-recorded meetings, lectures, and interviews.

Custom Summary Templates

Shape your summaries the way you work. Pick from 6 built-in templates or create your own.

Built-In Templates

Template	Best for
Standard Meeting	General meetings with agenda, decisions, and action items
Daily Standup	Quick status updates - yesterday, today, blockers
Project Sync	Cross-team alignment with milestones and dependencies
Retrospective	Sprint retros - what went well, what to improve
Psychiatric Session	Clinical documentation with session notes structure
Sales / Marketing Call	Pipeline updates, objections, next steps

Create Your Own

Section-level editor - Add, remove, and reorder sections with live markdown preview
Template CRUD - Create, edit, duplicate, delete, and export templates
Per-template icons - Pick an icon to identify your templates at a glance
Table format support - Templates can output summary sections as tables

This was F-10 in our prioritization doc - ranked HIGH for summary customization. Combined with F-11 (key action items), templates give you full control over what your AI summary captures.

Redesigned Meeting View

The meeting details page has been rebuilt from the ground up.

Tabbed interface - AI Summary, Transcript, and Preferences live in clean tabs instead of a cluttered two-panel layout
Floating ActionController - Post-stop processing controls float above the content for quick access to summarize, retranscribe, and export
In-page provider configuration - Change your summary provider without navigating away from the meeting
Atomic component system - New atoms/molecules/organisms architecture for consistent UI across the app

Smarter Local AI

Qwen 3.5 replaces Gemma as the default built-in summary model. It runs on half the RAM with better output quality.

Change	Before (1.1.1)	After (1.5.0)
Default model	Gemma	Qwen 3.5
Minimum RAM for 4B model	16 GB	8 GB
Built-in catalog	Multiple models	3 focused models (qwen3.5:2b, qwen3.5:4b, gemma3:4b)
Model hosting	HuggingFace only	S3 primary + HuggingFace fallback
Sampling	Generic presets	Benchmark-backed per-model presets

Lower barrier to entry - Users with 8 GB RAM can now run the 4B parameter model, previously requiring 16 GB
Deterministic summaries - Greedy sampling params (temperature: 0, top_k: 1) for consistent output with Ollama
Trimmed catalog - Removed unsupported models that caused confusion. If your saved model selection is no longer supported, the app blocks it with a clear message

FFmpeg Bundled

FFmpeg is now bundled as a sidecar binary. No more asking users to install it manually.

Build-time bundling - Platform-specific FFmpeg binary downloaded during cargo build
3-tier detection - Bundled binary first, then dev path, then runtime fallback
CI caching - GitHub Actions cache for faster builds
Hidden console - No more CMD window flash on Windows during media import

This directly eliminates a first-time setup friction point that blocked audio import and playback on machines without FFmpeg installed.

Bug Fixes

Fix	Impact
4B+ model loading (`gguf_init_from_file_impl: failed to read magic`)	Critical - models over 4B params wouldn't load
Iframe model chooser fallback with 8s timeout	High - offline users got stuck on model selection
Ollama cloud model parsing (`modelfile` field)	Medium - cloud-hosted Ollama instances failed
Thinking/reasoning token suppression	Medium - Ollama models wasted tokens on chain-of-thought
`whisperModel: null` serde crash in settings	Medium - corrupt settings could crash the app
FFmpeg console window on Windows	Low - cosmetic but distracting

Technical Details

Component	Change
Version	1.1.1 → 1.5.0
Development period	13 days (Mar 12 - Mar 24)
PRs merged	15
Files changed	158
Lines added	16,559
Lines removed	2,074
New Rust modules	4 (import.rs, retranscription.rs, decoder.rs, common.rs)
Built-in templates	6
Supported import formats	10
Default summary model	Qwen 3.5
Minimum RAM (4B model)	8 GB (was 16 GB)
New tests	13
Contributors	3

Prioritization Mapping

This release addresses several items from our user feedback prioritization(opens in new tab):

Priority Item	Status
F-6: Import Audio and Transcribe (HIGH)	Shipped
F-10: Templates (HIGH)	Shipped
F-11: Key Action Items (MEDIUM)	Shipped (via template sections)
B-2: Llama Helper / Sidecar Crashes (HIGH)	Fixed (llama-cpp-rs upgrade)

All Feature Guides

Audio Import - Import files by drag-and-drop
Retranscription - Re-process with a different model
Summary Templates - Built-in and custom templates
Meeting Details - Redesigned tabbed interface
Provider Configuration - Configure AI providers in-page
Share & Export - Export to PDF, DOCX, Markdown
Search & Replace - Find and replace across transcripts

Upgrade

Seamless upgrade from v1.1.1. All existing settings, recordings, and models are preserved. New features activate automatically.

The app will prompt you to update via the in-app update dialog. Click "Update Now" to install.

Download

Download Meetily Pro(opens in new tab)

Release Highlights