2 tokens/s decode and 40 tokens/s prefill
https://argonautlabs.ai/
Comments URL: https://news.ycombinator.com/item?id=50401869
Points: 2
# Comments: 0
AI & ML
Source：https://news.ycombinator.com/item?id=50401869
Kimi K3.1 locally on one GPU
Hacker News New0 views
PreviousType: Agent orchestrator for running Claude Code and Codex on cloudNextAxon invests $1.2B in Skydio, whose autonomous drones are used by public safety agencies, the US military, and 32 allied nations, at a $5.2B pre-money valuation (Carly Page/Resilience Media)
Related posts
Layer Rename & Cleanup: batch rename and delete empty layers in Aseprite (free Lite)
0dev.to
Open-Weight LLM vs Hosted API: What Self-Hosting Gives You That an API Can’t
0dev.to
Aizawl to Guwahati Flight Prices: Key Takeaways From a Booking-Window Analysis
0dev.to Data
Thinking Machines Ships Inkling 2 Small While Raising a 5B Round at 40B Valuation
0dev.to News
Ant Group Launches Ling-3.2-flash, a Native Multimodal Model
0dev.to News
OpenAI's $1.25 Trillion Funding Round Falls Short of Its $1.4 Trillion Target
0dev.to News