faster-qwen3-tts brings quantized Qwen3-TTS to Apple Silicon via GGML
andimarafioti · x · 2026-09-25
The open-source faster-qwen3-tts project now supports Apple Silicon via GGML, enabling real-time local speech generation with quantized Qwen3-TTS weights on Macs. It offers streaming and non-streaming modes: the Torch backend uses CUDA Graph (NVIDIA GPU required), while the GGML backend supports CUDA and Metal (macOS 14+). Install with pip; native Apple Silicon Python auto-selects the Metal runtime. 1.4k stars on GitHub.
More from Infra
- Altman says OpenAI aims to replace Nvidia in the medium to long term — zephyr_z9 · 2026-11-11
- Chinese University Bought Nvidia H200 Chips Despite US Blacklist — pstAsiatech · 2026-11-11
- Baidu's ERNIE Evolve tops public bests on NanoGPT Speedrun, brings RSI to real inference optimization — xiaohu · 2026-11-11
- Baidu says RSI auto-optimized ERNIE inference, 6.24x decode speedup on Kunxin P800 — xiaohu · 2026-11-11
- Nvidia's supply commitments jump from $119B to $279B in one quarter, locking up 2027-28 memory — tengyanAI · 2026-11-11
- Train 27B models on a 24GB GPU: ternary optimizer cuts memory to 1 byte/weight — oguzeldereli · 2026-11-11