ElevenLabs Pro - Account Info

Every paid node in this pack quietly spends your ElevenLabs credits. AccountInfo is the one that tells you how much runway you have left before the next top-up. It queries your account and returns your subscription tier, your current character usage, and your character limit - the kind of thing you check once, realize you're about to hit the cap mid-render, and save yourself a failed batch.

It's a config node, not a generation node, so the trade is different: it costs you a cheap authenticated GET request instead of credits. Under the hood it hits ElevenLabs' /v1/user/subscription endpoint by default, which returns your subscription dict directly. There's an endpoint dropdown with a user option too, but you almost never want it - more on that in a second.

Inputs are just api_key and that endpoint choice. Outputs are the useful part:

• subscription_tier - a string like "creator" or "free".
• character_count / character_limit - numbers you can actually wire into math nodes for a "90% used" warning light.
• account_json_redacted - the full account payload as pretty-printed JSON.

Now the part that should matter to you as much as the quotas: this node exists because v2.0 of the pack returned the raw /v1/user JSON, which includes your email, name, and billing IDs. And ComfyUI saves workflow state into the PNG metadata when you export an image. So anyone who shared one of those workflows was shipping their identity out with it. v2.1 switched the default endpoint to the PII-free subscription endpoint and added redact_account_pii(), which deep-strips email, names, and billing fields before anything is returned. The endpoint="user" option still exists for people who genuinely need the full record, but the tooltip flags it, and you should know it leaks if you go down that path.

This node also deliberately re-executes on every queue (it uses AlwaysExecuteMixin rather than the input-hash caching the paid nodes use). That's intentional - the GET is cheap and the whole point is a live reading, not a cached one.

Install is the same as the rest of the pack:

cd ComfyUI/custom_nodes
git clone https://github.com/IxMxAMAR/ComfyUI-ElevenLabs-Pro.git
pip install -r ComfyUI-ElevenLabs-Pro/requirements.txt

Then restart ComfyUI (or grab it from ComfyUI Manager under "ComfyUI-ElevenLabs-Pro"). Dependencies are just requests and soundfile; nothing to download.

One honest limitation: character_count measures characters, and different models bill credits at different rates (the Cost Estimator node tracks that, with flash/turbo lines at 0.5 credits per char versus 1.0 for the big multilingual models). So the count tells you usage, not exactly what the next batch costs. Treat it as a fuel gauge, not an invoice. And if you're about to run a big batch and want to know whether your tier even supports the mp3_44100_192 or opus formats you picked, this node is the fastest way to confirm which tier you're actually on.