Telegram: Send Video Album

If you've ever queued five video renders and watched Telegram get spammed with one message per clip, you know why this node exists. Telegram: Send Video Album is the gather-then-group version of Telegram: Send Video: it collects up to ten VIDEO outputs and delivers them as a single Telegram album (sendMediaGroup) - one swipeable message with all your clips inside, instead of a notification storm.

It's the thinnest node in the pack, but it's the one you want for actual batch work: variation renders, A/B comparisons, or dumping a whole sequence's clips to the same chat. Everything about the video handling is inherited from Telegram: Send Video, so if you've used that, there's zero new logic to learn here.

How it works

Each of the optional video_1 through video_10 inputs takes a VIDEO value and gets normalized the same way the single-video node does: a Comfy video object with .save_to(path), a dict with a path, a dict with frames + fps (encoded to MP4 via imageio / imageio-ffmpeg), a plain string path, or a single-element list containing a path. Everything becomes a real .mp4, then the clips are grouped into albums of up to 10 - Telegram's cap - and sent with sendVideo for inline previews. Temporary files are cleaned up afterwards.

If only one video ends up connected, it degrades to a single sendVideo rather than a pointless one-item album. The caption sits on the first item of the first album, same as the image album.

The inputs that matter

• video_1 … video_10 (all optional VIDEO) - wire up to ten video sources; unconnected inputs are ignored.
• caption (STRING, multiline) - message text on the first item.
• send_as_video (boolean, default True) - the only toggle. It's on by default and that's fine; there's no document mode here at all.

No outputs - it's an output node, so it ends the workflow.

Installing it

ComfyUI Manager, search "TelegramSender". Or:

cd ComfyUI/custom_nodes
git clone https://github.com/smbdsbrain/ComfyUI-TelegramSender

Then the real dependencies, in ComfyUI's Python environment:

pip install requests imageio imageio-ffmpeg

imageio-ffmpeg is required for any input that arrives as frames rather than a ready file. Restart ComfyUI, node under utils/telegram.

Credentials are shared across the pack, first match wins: ComfyUI Settings → Telegram Sender → Credentials (recommended), a telegram_config.json next to the node, or the TELEGRAM_BOT_TOKEN / TELEGRAM_CHAT_ID env vars. Token from @BotFather, chat ID numeric or @channel.

Troubleshooting

• It's slow, and it's all slow - each video is normalized and encoded sequentially before anything sends. Several long clips can mean a long wait; that's the tradeoff for one clean message.
• Caption only on the first clip - expected, that's how Telegram albums work.
• More than 10 videos - split into multiple albums automatically.
• "imageio missing" on a frames input - you skipped the pip install of imageio / imageio-ffmpeg.
• "bot_token and/or chat_id not specified" - the pack's standard credential error, pointing you at Settings.

If you want the raw files as downloadable documents rather than inline video - this node has no document mode - keep a Telegram: Send Video per output, or just accept the album preview.