theserverless.dev InferenceList

Models

Showing 1–4 of 4 · Sorted by input price, low to high.

More filters
Model Provider Modality Input ↑ Output Context
NVIDIA: Nemotron 3 Nano Omni (free) NVIDIA text audio image video $0/1M $0/1M 256K
NVIDIA: Nemotron 3.5 ASR Streaming Multilingual 0.6B NVIDIA audio Not published— Not published— —
NVIDIA: Nemotron 3.5 Transcribe NVIDIA audio Not published— Not published— —
NVIDIA: Parakeet TDT 0.6B v3 NVIDIA audio Not published— Not published— —

Token rates are USD per 1M tokens. A dash means the source did not publish that amount. A zero is a real zero.