Hi, I have a question, and I'd appreciate it if an...
# inferenceg
gentle-traffic-46145
11/12/2025, 7:26 PMHi, I have a question, and I'd appreciate it if anyone knows the answer ! 🙂
For GPT 4.1, the prompt can be cached for $0.50, but is it the same with LiveKit inference? The pricing lists input and output, but not caching.
cc @rough-gpu-50664
r
refined-appointment-81829
11/12/2025, 7:41 PM@gentle-traffic-46145 Can you do me a favor and please thread your messages? Otherwise it gets confusing for which on folks should reply to.
refined-appointment-81829
11/12/2025, 7:41 PMHere is the other part of the question that was in the main channel:
-----
Nicolas Raulin [1:38 PM]
Usage: UsageSummary(llm_prompt_tokens=171161, llm_prompt_cached_tokens=158720, llm_input_audio_tokens=0, llm_completion_tokens=1858, llm_output_au>
[1:39 PM] Is the cached prompt paid at $2 per entry or at $0.50 as indicated in the Gpt doc for 4.1?
👍 1
g
gentle-traffic-46145
11/12/2025, 7:42 PMSorry for the inconvenience, thank you !!
Do you have the answer to your question ?
r
rough-gpu-50664
11/12/2025, 7:51 PMLiveKit Inference currently does not discount cached tokens. We're working on enabling this feature.
g
gentle-traffic-46145
11/12/2025, 7:54 PMOkay great, thanks for the quick reply ! :)