开启时会模糊预览图，关闭后正常显示
RT @justinblau: This is the cycle we will finally see the rise of non-extractive creator products. Anti-slop, transparent data, direct con…
Building @joinbond | prev @a16zcrypto | programmer | magician
Origin: Estonia Accuracy: 56% Size: baguette Range: 2 km «Although he does not give a price for the weapon, he says it comes in at roughly a tenth of existing rocket air defence systems (a Stinger missile markets for around £400,000)» status: ngmi Just give them money, losers
man do I despise Eurotrash that are grifting off this tragedy even as they feign concern for their own safety. only ever hustling to fill their pockets. Suicidal, almost Russian Army levels of scumminess it's a shame that Putin is so obsessed with killing Slavs specifically
RT @steventey: October was the largest month in @dubdotco history. In October, we made more revenue than all of 2024 combined 🚀 Excited t…
Founder of https://t.co/yyLfH8mOar and https://t.co/ZzTStsMvdh
just to save face, the main take here is we'll have AGI ~6 months after a model can reproduce a correct, working GPT-2 clone in plain C, with any tweaks you ask. we all agree LLMs will eventually do that, right? ofc, if they take 10 years to do so, then we're that from AGI
Kind / Bend / HVM / INets / λCalculus
小红书发的一款具备工具调用能力的多模态模型：DeepEyesV2，会自己找工具用，能看图、写代码、搜网页、组合工具解决复杂问题 任务自适应，可以根据当前任务性质智能选择最合适的工具；也可以组合多个工具进行解决复杂问题 可以根据当前上下文信息，选择性的决定是否需要调用工具以及调用哪个工具 在FVQA-test、MMSearch、SimpleVQA上超GPT-4o、Gemini 2.5 Pro，以及其他搜索模型 #DeepEyesV2 #AgenticMLLM
github：https://t.co/zoGLnLMTho
Speedtest Tracker 是一个基于 Laravel 的自托管仪表板，利用 Speedtest CLI 定期测速、记录性能与延迟，并通过历史数据帮助发现 ISP 的实际表现，支持 Docker 部署，便于长期运行与数据保存。 https://t.co/bFEQiHgwQd
🧠在家居士 | 🥦素食者 | 🏃🏻马拉松爱好者 | 💰省钱小能手 | 搭🪜技术资深学者 | 👨💻科技宅 | 🆕更新狂 | 🆅 六边型战五渣
