Even 'uncensored' models can't say what they want
(news.ycombinator.com)
1.
2.
We got 207 tok/s with Qwen3.5-27B on an RTX 3090
(news.ycombinator.com)
3.
Kimi K2.6: Advancing open-source coding
(news.ycombinator.com)
4.
5.
Running local models on Macs gets faster with Ollama's MLX support
(arstechnica.com)
6.
LLM Neuroanatomy II: Modern LLM Hacking and Hints of a Universal Language?
(news.ycombinator.com)
7.
Flash-MoE: Running a 397B Parameter Model on a Laptop
(news.ycombinator.com)
8.
MacBook M5 Pro and Qwen3.5 = Local AI Security System
(news.ycombinator.com)
9.
Qwen3.5-397B at 4.74 tok/s using 5.9GB RAM
(news.ycombinator.com)
10.
Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers
(news.ycombinator.com)
11.
Qwen3.5: Towards Native Multimodal Agents
(news.ycombinator.com)