Manual Model Download

What is Meetily downloading?

When you install Meetily for the first time, onboarding has exactly two downloads. You'll see them labeled on screen as "Step 1" and "Step 2":

If both screens got stuck, you'll do this twice. Otherwise, just the one that failed.

Which screen is stuck?

Look at the title and subtitle of the onboarding screen you're stuck on, then jump to the matching section below.

Step 1: Find Meetily's models folder

Open the folder below in Finder (Mac) or File Explorer (Windows) and keep it open — you'll drop files into it later. You only need to do this once, even if you have to do both Section A and Section B.

Section A: Transcription model (Parakeet)

Skip this section if Step 1 already finished and only Step 2 is stuck.

A.1: Download these 4 files

Right-click each link and choose "Save Link As…" (Chrome / Edge / Safari) or "Save Target As…" (Firefox). Don't left-click — your browser will try to open a 622 MB file as a webpage.

You can save the files anywhere convenient first (your Downloads folder is fine) and move all four into the Parakeet folder in one go after they finish — no need to save directly into Application Support.

The encoder is the big one — give it a few minutes. The other three finish in seconds.

A.2: Put the files in the right place

Inside the models folder you opened in Step 1:

1. Create a folder called parakeet (all lowercase).

2. Open parakeet. Inside it, create another folder with this exact name (copy-paste to avoid typos):

   Code language: text

   parakeet-tdt-0.6b-v3-int8

3. Drop all 4 downloaded files into that folder.

When you're done, the structure should look like this:

pro.meetily.ai/
└── models/
    └── parakeet/
        └── parakeet-tdt-0.6b-v3-int8/
            ├── encoder-model.int8.onnx           (~622 MB)
            ├── decoder_joint-model.int8.onnx     (~17 MB)
            ├── nemo128.onnx                      (~140 KB)
            └── vocab.txt                         (~94 KB)

MODEL_DIR="$HOME/Library/Application Support/pro.meetily.ai/models/parakeet/parakeet-tdt-0.6b-v3-int8"
BASE_URL="https://huggingface.co/istupakov/parakeet-tdt-0.6b-v3-onnx/resolve/main"
mkdir -p "$MODEL_DIR"
 
curl -L --fail -o "$MODEL_DIR/encoder-model.int8.onnx"        "$BASE_URL/encoder-model.int8.onnx"
curl -L --fail -o "$MODEL_DIR/decoder_joint-model.int8.onnx"  "$BASE_URL/decoder_joint-model.int8.onnx"
curl -L --fail -o "$MODEL_DIR/nemo128.onnx"                   "$BASE_URL/nemo128.onnx"
curl -L --fail -o "$MODEL_DIR/vocab.txt"                      "$BASE_URL/vocab.txt"

$modelDir = Join-Path $env:APPDATA "pro.meetily.ai\models\parakeet\parakeet-tdt-0.6b-v3-int8"
$baseUrl  = "https://huggingface.co/istupakov/parakeet-tdt-0.6b-v3-onnx/resolve/main"
New-Item -ItemType Directory -Force -Path $modelDir | Out-Null
 
$files = @(
  "encoder-model.int8.onnx",
  "decoder_joint-model.int8.onnx",
  "nemo128.onnx",
  "vocab.txt"
)
foreach ($f in $files) {
  Write-Host "Downloading $f ..."
  Invoke-WebRequest -Uri "$baseUrl/$f" -OutFile (Join-Path $modelDir $f)
}

Section B: Summary model (Qwen 3.5)

Skip this section if Step 2 already finished and only Step 1 is stuck.

B.1: Pick the right variant

Your Step 2 screen tells you which one you need. Match the title:

B.2: Download the file

Right-click the link for your variant and choose "Save Link As…"

Save it with the exact filename shown above. Don't rename it — capital Q, capital K, capital M all matter.

If you're not 100% sure which variant Meetily picked for you, save both files into the summary folder. Meetily uses only the one it's set for; the other just sits there unused, no harm done.

B.3: Put the file in the right place

Inside the models folder you opened in Step 1:

1. Create a folder called summary (all lowercase).
2. Drop the .gguf file directly inside.

pro.meetily.ai/
└── models/
    └── summary/
        └── Qwen3.5-2B-Q4_K_M.gguf       (or Qwen3.5-4B-Q4_K_M.gguf)

SUMMARY_DIR="$HOME/Library/Application Support/pro.meetily.ai/models/summary"
mkdir -p "$SUMMARY_DIR"
 
# Default: Qwen 3.5 2B (~1.27 GB)
MODEL_FILE="Qwen3.5-2B-Q4_K_M.gguf"
HF_REPO="unsloth/Qwen3.5-2B-GGUF"
# For Qwen 3.5 4B (~2.6 GB):
# MODEL_FILE="Qwen3.5-4B-Q4_K_M.gguf"
# HF_REPO="unsloth/Qwen3.5-4B-GGUF"
 
curl -L --fail -o "$SUMMARY_DIR/$MODEL_FILE" "https://huggingface.co/$HF_REPO/resolve/main/$MODEL_FILE"

$summaryDir = Join-Path $env:APPDATA "pro.meetily.ai\models\summary"
New-Item -ItemType Directory -Force -Path $summaryDir | Out-Null
 
# Default: Qwen 3.5 2B (~1.27 GB)
$modelFile = "Qwen3.5-2B-Q4_K_M.gguf"
$hfRepo    = "unsloth/Qwen3.5-2B-GGUF"
# For Qwen 3.5 4B (~2.6 GB):
# $modelFile = "Qwen3.5-4B-Q4_K_M.gguf"
# $hfRepo    = "unsloth/Qwen3.5-4B-GGUF"
 
Write-Host "Downloading $modelFile ..."
Invoke-WebRequest -Uri "https://huggingface.co/$hfRepo/resolve/main/$modelFile" `
  -OutFile (Join-Path $summaryDir $modelFile)

Step 2: Quit and restart Meetily

Meetily only scans the models folder when it starts up. You have to fully quit and reopen — closing the window isn't enough.

After restart, the onboarding screen should show a green checkmark ("Transcription Model Ready" / "Summary Model Ready") instead of the progress bar. You can click Next and continue.

If you'd already finished onboarding when you started this guide, no restart action is needed in onboarding — just open Settings in Meetily. The transcription model shows as Downloaded under Transcript, and the summary model under Built-in AI Models.

Troubleshooting

Related

• Pluggable Transcription — how to switch between Parakeet, Whisper, and cloud providers once your model is installed.