‹ All stories

Whistle

Product · 1 update

Where it stands

As of Oct 11, Whistle is a single 16.9 MB .cact file that runs on the same CPU Needle engine with no GPU: English, German, French, Spanish, Italian, Dutch and Polish, word timestamps, keyword biasing, and speech embeddings, with ~11 ms to first token on Apple Silicon.

Timeline · newest first
  1. Official

    Cactus Compute releases Whistle, a 16.9 MB on-device speech-to-text model for seven languages

    Cactus Compute published this on Oct 11, 2026 (source: cactuscompute.com).

    Whistle is a single 16.9 MB .cact file that runs on the same CPU Needle engine with no GPU: English, German, French, Spanish, Italian, Dutch and Polish, word timestamps, keyword biasing, and speech embeddings, with ~11 ms to first token on Apple Silicon. Weights are on Hugging Face; the engine ships for seventeen platforms including mobile and WebAssembly.

    Source: Cactus Compute
Related stories
NVIDIA36 updates · Models→Claude33 updates · Models→Google Gemini28 updates · Models→Qwen21 updates · Models→
How we label
Official
An institution announces something that has already happened, through its own channel or a founder or executive account.
Statement by …
A named person's views, predictions, plans or comments, including remarks reported by media when the original is unavailable.
Reported by …
Only media or another third party is saying it. We name the outlet.
Unconfirmed
Circulating online (posts, comments, leaks). No company, named person or outlet has confirmed it.

Items older than 72 hours never go in Top stories and show their original date.

DailyCitedRSSAboutPrivacy