Because if you don't offer backups your customers will chose someone else who does
@simonw
-
Running Coding Agents Safely: Sandbox Execution Without Skip Permissions
By
–
I can't imagine any way of running a coding agent for 8 hours without skip perms – the trick is to run it in a sandbox where mistakes don't matter
-
Talkie AI tool created by Status Effects and collaborators
By
–
talkie was created by @status_effects and @DavidDuvenaud and @AlecRad
-
AI Image Generation Limitations and Era-Appropriate Responses
By
–
It didn't quite manage to draw me a pelican riding a bicycle, but I still appreciated its era-appropriate response
-
Talkie: New Vintage Language Model Trained on Historical Pre-1931 English
By
–
Some notes on talkie, a new "vintage language model" from a team including Alec Radford (yes, that Alec Radford) "trained on 260B tokens of historical pre-1931 English text"
-
Comparing Parakeet MacWhisper MLX Audio VibeVoice Tools
By
–
I've been using Parakeet via MacWhisper, haven't run a direct comparison with MLX audio and VibeVoice
-
MLX Audio STT Tool Command Line Interface
By
–
Good catch, that works fine too: uv run –with mlx-audio mlx_audio.stt.generate –help
-
MLX one-liner downloads runs VibeVoice ASR model locally
By
–
Here's a uv one-liner that downloads and runs the MLX model against a local mp3 file uv run –with mlx-audio python -m mlx_audio.stt.generate
–model mlx-community/VibeVoice-ASR-4bit
–audio lenny.mp3 –output-path lenny
–format json –verbose –max-tokens 32768 -
VibeVoice: MIT-Licensed Speech-to-Text with Diarization
By
–
Microsoft's MIT licensed VibeVoice speech-to-text model (think Whisper with speaker diarization) is really good – my notes on running the 5.71GB 4bit MLX conversion on an M5 MacBook, using about 60GB of RAM at peak and transcribing 1hr of audio in ~9 mins
