Talk - Talk

12 [ai]What open-source LLM are you running locally in 2026? I switched from Llama 3 to Gemma 4 last week a by techfan 9h | 2 replies

15 [ask]Ask: What is the cheapest way to serve a 7B model for production use? I need about 1000 requests per by curious 14h | 1 replies

8 [dev]Hot take: most AI wrapper startups will fail not because of technology but because they have no dist by builder 11h | 0 replies

5 The pace of open model releases in 2026 is insane. We went from GPT-4 being unreachable to multiple by observer 20h | 0 replies

3 [show]Show: I built a CLI tool that benchmarks any GGUF model on your hardware in 60 seconds. Tests throug by maker 1d | 0 replies