Keep on to blur preview images; turn off to show them clearly
![The next evolution of image generation is here. Workers AI is hosting the new FLUX.2[dev] by @bfl_ml
Featuring:
✨Multi-reference image inputs for consistent characters and iterative image generations
✨Real-world knowledge informing lighting & spatial logic for coherent scenes
✨Web design and infographic capabilities to build digital assets
Start generating @cf/black-forest-labs/flux-2-dev](/_next/image?url=https%3A%2F%2Fpbs.twimg.com%2Fprofile_images%2F991779658147377152%2FESp67BNN_400x400.jpg&w=3840&q=75 "The next evolution of image generation is here. Workers AI is hosting the new FLUX.2[dev] by @bfl_ml
Featuring:
✨Multi-reference image inputs for consistent characters and iterative image generations
✨Real-world knowledge informing lighting & spatial logic for coherent scenes
✨Web design and infographic capabilities to build digital assets
Start generating @cf/black-forest-labs/flux-2-dev")
The next evolution of image generation is here. Workers AI is hosting the new FLUX.2[dev] by @bfl_ml Featuring: ✨Multi-reference image inputs for consistent characters and iterative image generations ✨Real-world knowledge informing lighting & spatial logic for coherent scenes ✨Web design and infographic capabilities to build digital assets Start generating @cf/black-forest-labs/flux-2-dev
Read our blog: https://t.co/yyFJf824Bo
Moats before (gross) margins. True five years ago, and true today. https://t.co/cxkM4N78m6
Software is eating the world, but I’m eating @saltandstraw. My initials were AI before it was cool. Partner @a16z growth fund.
RT @HuggingPapers: Microsoft & Oxford Research ask: Are your UIs ready for agents? Introducing a breakthrough framework where Computer-Use…
AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo ,submit papers here: https://t.co/UzmYN5XOCi
RT @falkordb: HAPPENING TOMORROW: Text-to-SQL/Cypher Workshop. We'll demo our approach to txt2SQL/Cypher, and feature a #dbt project from…
VC by day @untappedvc, builder by night: @babyagi_, @pippinlovesyou @pixelbeastsnft. Build-in-public log: https://t.co/UdHHGbZba5
Not even Japanese nationalists can hate Wuyue Fascinating kingdom, seemingly nobody hated it
We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1
最近玩疯了。 这两周基本是在 Gemini 3 和 Nano Banana Pro 之间来回折腾，跟被点满了生图天赋一样，天天生成到天亮。 模型强大带来的效果丝滑，很容易被这种爽感裹挟进去。 但可能正因为玩得太嗨，反倒让我注意到一个反差： 大家好像都把 agent 暂时忘了。 不知道有多少人跟我一样，时间线被更强的模型和更丝滑的生图刷得密不透风。 agent 那一套跑流程的讨论从时间线里消失了。 这很值得警惕。最近在其他平台在看一些项目、也跟几家公司聊，能感觉 agent 在悄悄蓬勃起来。 不吵不闹，也没有那种炸裂发布会。但能看到它开始钻进各种真实业务里。 以前要人盯着的小任务，它们有的已经开始自己跑了。 模型升级确实很爽，但 agent 的演化到了没声量但稳稳前进的时刻。 警醒下自己， 玩模型当然好爽，但如果一直停在好玩，很容易忽略掉一个事实： 最后决定业务价值的，往往是那些能把模型用起来的流程和系统。 所以也给自己按了个暂停键： 收敛一下，该回到 agent 这条线索上继续深挖了。 有时候，冷门不是真的没人做，只是热闹把声音盖住了。 而真正值得跟的方向，往往就是这种安静并且一步一个脚印的演进。
行道途中。非求速成，惟求通达。 2023 年扎进AI ，打通Know-How，不少赚钱项目，踩过坑，也见过光。 围城里待得够久了，出来聊聊世界，聊聊技术、聊聊赚钱。