Paris Bench 🇫🇷 - DeepSeek 4.1 Flash Max
Ista IQ2_XS · via Project Maya · 33 tok/s ⚡ Nice speed, but the quality loss from quantization is noticeable. Excited for Qwen4! 🤞 Smaller models would let us run higher-quality quants ✨ cc PeasantSmith
Hedge fund manager and quant trading author Ernie Chan on why hedge funds can't just copy OpenAI and DeepSeek
"In theory, everybody has access to this." "Everybody has read OpenAI and DeepMind's paper. Why can't they become the next Google?" "It's not that easy." "We have a
还的是Qwen4！ Loktar的这段森林中雨夜的寺庙场景，Qwen4 116k token ，DeepSeek Harness 732k token !!!
差了接近6倍多7倍！！ 很多人已经迫不及待地想在本地运行Qwen4了！ 在我看来，Qwen4的3D建模和图形方面的性能和性价比远超其他开源模型！ 不知道打的过Opus 5.5不？
DeepSeek Harness: an open-source AI agent where everything is a plugin, even the interface.
You build it, you don't rent it. Full Short:
You have reached the end of the archive
All of deepseek