Nodes/Elevenlabs-ComfyUI/ElevenLabs History
ComfyUI Node

ElevenLabs History

Audit what your API key has generated

By karthikg-09·Created 11 months ago·Updated 11 months ago· 1
ElevenLabs History
    • history
    api_key
    page_size10
    refreshfalse

    Every TTS call you make against ElevenLabs lands in your account's generation history - even the ones you made from outside ComfyUI. This node reads that ledger back and prints it as readable text. It's not a creative tool. It's the "where did my credits go" node, and once you're burning real money on TTS volume, that's exactly the question you want answered fast.

    How it works

    It calls ElevenLabs' /v1/history endpoint with your page_size and formats the response into a text block: each entry shows the first 50 characters of the generated text, the voice used, the date, and how many characters it cost you. Straight GET, no audio returned, nothing downloaded - this is pure telemetry.

    Inputs

    • api_key - your key.
    • page_size - 1 to 100, default 10. How many recent items to pull.
    • refresh - a boolean, default false. It's accepted but effectively cosmetic: every run hits the live endpoint anyway, so you don't need to flip it to get fresh data.

    Output: a history STRING - the formatted list, or "No history items found" on an empty account. It's also printed to the ComfyUI console, which is honestly the more convenient place to read a long list.

    Install

    Same pack, same command:

    cd ComfyUI/custom_nodes
    git clone https://github.com/karthikg-09/Elevenlabs-ComfyUI.git
    cd ElevenLabs-ComfyUI && pip install -r requirements.txt
    

    or ComfyUI Manager → "Elevenlabs-ComfyUI" → restart. Add Node → ElevenLabs.

    What it's actually good for

    Three things, in decreasing order of excitement. First, credit forensics: if your quota vanished, this shows you exactly which runs ate it, with per-item character counts. Second, workflow debugging: when a TTS node "worked" but the output surprised you, the history entry tells you which voice and what text actually got sent. Third, recall: need to re-run something you generated last week? The history shows the prompt and the voice, so you can reconstruct it.

    The honest limit: it's a text audit, not a library. There's no audio preview, no download button, no way to grab the generated file back into your graph from here. For that you'd hit the ElevenLabs dashboard or the API's history-item audio endpoint yourself. Within those bounds it's a genuinely useful accounting node - the kind of thing you don't think you need until you're staring at a drained credit balance.

    CategoryElevenLabs

    Inputs (3)

    NameTypeDefaultDescription
    api_keySTRING
    page_sizeINT101–100
    refreshBOOLEANfalse

    Outputs (1)

    NameTypeDescription
    historySTRING