DEV Community

devs profile picture
Mistral Large 4 Open Weights: Running a 1T MoE Locally Without the Cloud Tax

Mistral Large 4 Open Weights: Running a 1T MoE Locally Without the Cloud Tax

Comments
5 min read

Want to connect with devs?

Create an account to connect with devs. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
Hugging Face AnySpeechRecognizer: swap Apple, Parakeet or Qwen3-ASR without rewriting your STT path

Hugging Face AnySpeechRecognizer: swap Apple, Parakeet or Qwen3-ASR without rewriting your STT path

Comments
4 min read
Ollama 0.40.13 puts Liquid LFM3 audio on Windows: local STT/TTS that fits in 2 GB

Ollama 0.40.13 puts Liquid LFM3 audio on Windows: local STT/TTS that fits in 2 GB

1
Comments 1
5 min read
NVIDIA Nemotron 3.5 Transcribe is offline on purpose: fitting a 4.3B-active speech LLM into a live Windows call path

NVIDIA Nemotron 3.5 Transcribe is offline on purpose: fitting a 4.3B-active speech LLM into a live Windows call path

Comments
6 min read
Realtime STT at $0.36/hour is not the hard part: turn-taking and the Windows capture path

Realtime STT at $0.36/hour is not the hard part: turn-taking and the Windows capture path

Comments
4 min read
BYOK for a live call assistant: why the app never sees your provider keys

BYOK for a live call assistant: why the app never sees your provider keys

1
Comments
1 min read
Storing users' API keys in a Windows desktop app with DPAPI, from Python, without a dependency

Storing users' API keys in a Windows desktop app with DPAPI, from Python, without a dependency

Comments
4 min read
Floating overlay that stays out of screen-share captures on Windows

Floating overlay that stays out of screen-share captures on Windows

Comments
1 min read
Building a dual-track audio pipeline on Windows: mic + system audio without joining the call

Building a dual-track audio pipeline on Windows: mic + system audio without joining the call

Comments
2 min read
Stop hammering a rate-limited LLM API: per-provider cooldowns in a live assistant

Stop hammering a rate-limited LLM API: per-provider cooldowns in a live assistant

2
Comments 5
4 min read
Your WAV file says it's empty: keeping recordings readable when the process gets killed

Your WAV file says it's empty: keeping recordings readable when the process gets killed

1
Comments
5 min read
Two streaming speech-to-text bugs that left no error: binary frames and keyterm budgets

Two streaming speech-to-text bugs that left no error: binary frames and keyterm budgets

1
Comments
4 min read
Product screenshots from the real UI, without real user data: Electron, a mock API and capturePage

Product screenshots from the real UI, without real user data: Electron, a mock API and capturePage

Comments
3 min read
Serving Markdown to AI agents from a Cloudflare Worker, without a second copy of the site

Serving Markdown to AI agents from a Cloudflare Worker, without a second copy of the site

1
Comments
4 min read
Capturing the mic and the call as two separate tracks on Windows with cpal and WASAPI loopback

Capturing the mic and the call as two separate tracks on Windows with cpal and WASAPI loopback

Comments
5 min read
Where the time goes in a live AI call assistant: my code takes 150 ms, the model takes the rest

Where the time goes in a live AI call assistant: my code takes 150 ms, the model takes the rest

Comments
4 min read
loading...