Qwen 27B Compressed to Run on iPhone
pmttyji · reddit · 2026-07-13
PrismML announces it has compressed the Alibaba open-source model Qwen 3.6-27B to run on iPhone 17 Pro, with download planned for next week.
Details:
- Original model has 27 billion parameters, compressed to under 4GB (from 54GB) to fit on a phone.
- PrismML claims the model still handles complex chat, reasoning, autonomous agents, and software coding.
- The CEO believes most AI computing will shift to local devices, with cloud handling only the last portion of high-end needs, fundamentally changing AI economics.
- The technology stems from Caltech mathematical research, with patents exclusively licensed to PrismML.
- The article contrasts Apple's on-device models: Apple's models have more parameters but use sparse architectures, activating far fewer than PrismML's 'full-activation' approach.
Overall, this is a typical on-device large model compression / local inference story, also touching on capability and deployment economics.
Related event: Bonsai 27B: The first 27B model that runs on phones(15 posts)→
More from Infra
- Anthropic hits $16.3B H1 revenue as compute efficiency jumps 5.2x in 18 months — daniel_mac8 · 2026-10-27
- Tenstorrent runs Proteina protein model at high throughput, claims no HBM shortage — gajesh · 2026-10-27
- YC-backed Hopper launches voice-agent inference cloud: 80 ms TTFT at $0.30/1M tokens — ycombinator · 2026-10-27
- A100 rental prices still rising in 2026 per a16z data, Jevons paradox hits compute — VraserX · 2026-10-27
- Anthropic S-1 reveals $800B+ in off-books future compute commitments — GaryMarcus · 2026-10-27
- ~42% of Anthropic's ~11GW contracted compute is NVIDIA GPUs — ns123abc · 2026-10-27