Vendor page
Fluid Speech Mini (GGUF): model card on Hugging Face
FluidVoice Labs, Hugging Face, 2026
Read the originalRead on October 22, 2026
- What it is
- The model card for Fluid Speech Mini, an English speech-to-text model for dictation that FluidVoice Labs publishes on Hugging Face. Its Limitations section says the model is English only, and that the vocabulary of the Parakeet model it is built on covers 25 languages but other languages were not trained or measured. The page also describes the GGUF build's file size and its runtime on CPU and CUDA.
- How much weight it carries
- The maker's own model card, read from its Hugging Face page, which shows no date, so the read date is the only date. The language limit is a plain statement by the maker. The benchmark figures on the page are self-reported runs of the leaderboard's code, not entries on the public board, but this row does not use them.
Facts taken from this source
- The model card for Fluid Speech Mini lists English only as its limitation, so the model recognizes one language, English. Other languages from its base vocabulary were not trained or measured.
“Limitations English only. The vocabulary is Parakeet v3's 25-language one, so other languages may come out, but they were not trained or measured and are expected to be worse.” (Limitations)