Post-Cutoff

Cactus ComputeNeedle familyCurrentSpeech and audioReleased 2 Oct 2026

Cactus Whistle (16.9 MB on-device speech-to-text)

Launch: https://x.com/cactuscompute/status/2106083041265562075 · Related: phonon-2, whisper-large-v3.

Model idCactus-Compute/whistle

PricePrice not published

Checked3 October 2026

Where it stands

10newer speech and audio models since its release

The 5 newest are listed. All speech and audio models

Weights
Weights are public under a permissive licence (Apache-2.0).

How to call it

3 ways to use Cactus Whistle (16.9 MB on-device speech-to-text), with the model id each provider expects.

Providers, model ids, endpoints and docs for Cactus Whistle (16.9 MB on-device speech-to-text)
ProviderModel idEndpointDocs
Hugging Face Cactus-Compute/whistle huggingface.co
GitHub (engine) github.com
Browser demo cactuscompute.com

Pricing

Price not published.

No price source linked. Checked 3 October 2026.

Specs

Input
Audio
Output
Text
Context window
Not published
Max output
Not published
Open weights
Yes
Licence
Apache-2.0
Released
2 October 2026
Status
Current
Type
Speech and audio

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Speech recognition in a 16.9 MB file, on CPU

    One 16.9 MB file with no dependencies, running in the same C++ engine as Cactus’s Needle model; ~11 ms to first token; transcribes up to 30 s per pass in English, German, French, Spanish, Italian, Dutch and Polish with language detection, word timestamps and speech embeddings. Vendor benchmarks: ahead of Whisper base on most sets at ~9× smaller and ~6× faster.

    Source: cactuscompute.com

Notes

Aimed at phones, wearables, robots, smart home, cars and microcontrollers. Compared with Whisper base (a small 2022 model), not with large ASR models; vendor-run numbers. Launch post ~167k views; Nobara/GE-Proton maintainer GloriousEggroll reported it made his home assistant respond ‘SO damn fast’.