parakeet_cuda: local Parakeet transcription on a 2GB VRAM GTX 750 Ti, ~20x realtime
mgostIH · x · 2026-10-02
A developer wrote a custom CUDA runtime (parakeetcuda, Apache-2.0) so NVIDIA's Parakeet TDT 0.6B v3 and Nemotron diarization models run on a GTX 750 Ti with just 2GB VRAM. Exact FP32 inference with cuBLAS: 280 MiB for ASR, 384 MiB with diarization, roughly 20x realtime — local audio transcription on nearly any hardware.
More from Infra
- tandem.cpp v0.3.0: 2-6x faster prefill for models too big for VRAM, Maya DeepSeek-V4.1 support — smflx · 2026-11-16
- Developers poll real costs of running coding agents on remote VMs for work — radim11 · 2026-11-16
- Datadog's new gateway returns 402 when a model's budget runs out—so agents just retry — YvesMulkers · 2026-11-16
- Max-quality local Minimax H3 video on a 4090: community-tested int8 setup — newaccount47 · 2026-11-16
- Local V2V outfit transfer on a 16GB M3 MacBook Air: 15.7 min cut to 2.4 min — might_thinker · 2026-11-16
- FlashStates hits 3.14x vLLM prefill throughput for hybrid-architecture agentic inference — 机器之心 · 2026-11-16