I made a 35B AI model run at 29 tokens/sec on consumer GPUs.
Stock llama.cpp: 10 t/s
Baldur KSL: 29 t/s Same model. Same hardware. 3x faster. Free tier, no signup. Just download and run.
https://t.co/b35m3wT6GI #LocalLLM#AI#llama_cpp#MoE@_akhaliq@reach_vb@LocalLLaMA
I made a 35B AI model run at 29 tokens/sec on consumer GPUs.
Stock llama.cpp: 10 t/s
Baldur KSL: 29 t/s
Same model. Same hardware. 3x faster.
Free tier, no signup. Just download and run.
https://t.co/b35m3wT6GI
Demo:https://t.co/jU7JaEzrUb
@Qaborai @_akhaliq@reach_vb@LocalLLaMA
HUGE milestone: AI Terminal is now available on ALL platforms!
๐ง Linux โ
๐ macOS โ NEW!
๐ช Windows โ
Get AI-powered debugging in your terminal. No more copy-pasting errors to
ChatGPT.
Download: https://t.co/yrZZQRDMfO
#DevTools#AI#macOS#linuxcommands
Windows developers rejoice! ๐
AI Terminal now supports Windows. Run any command and get instant AI analysis when errors occur. No more copy-pasting to ChatGPT. Download: https://t.co/yrZZQREk5m
Demo: https://t.co/5IlH5QyUj4 #BuildInPublic#AI#DeveloperTools#linuxcommandline
๐ช Windows version coming TODAY!
Setting up Windows build environment right now - AI Terminal will be available for Linux, and Windows by end of day.#18 on @ProductHunt
Demo: https://t.co/5IlH5QyUj4
Download Linux now: https://t.co/uaQ66MTisl
#DevTools#AI#BuildInPublic
@elonmusk Perfect timing - just launched AI Terminal which brings AI to developers' terminals.
Adding Grok support soon - would love to have Grok explain terminal errors!
Demo: https://t.co/5IlH5QyUj4
I built AI Terminal to detect and explain terminal errors automatically.
See it in action (57 sec):
https://t.co/5IlH5QyUj4
Live on Product Hunt today!
https://t.co/SkcUc6biOt
#Developer#AI#Terminal
AI Terminal - Your terminal just got smarter
Automatically detects errors and explains them in plain English + suggests fixes Local AI (Ollama/LM Studio) or Cloud (GPT-4, Claude) Real-time detection Linux ready Free trial https://t.co/yrZZQRDMfO
#DevTools#AI#Terminal#Linux