[Notice] The DeepSeek online model has been upgraded to version V3.1, with the context length expanded to 128k. You are welcome to test it on the official website, APP, and mini-programs. The API call method remains unchanged.

The Qwen team has released its own image editing model: Qwen-Image-Edit It has two core functionalities: Powerful image content editing capabilities: It can perform "big edits," such as transforming a photo into a Miyazaki-style image or rotating the perspective of objects in the image, as well as "small edits," like removing a strand of hair or changing a sign in the background, all while ensuring that other parts of the image remain completely unchanged. Precise text editing capabilities: It allows direct modification of Chinese and English text within images, whether adding, deleting, or altering, while striving to maintain the original font and style. It can even correct typos in calligraphy works step by step, just like editing a document. In simple terms, this is an AI editing tool that can understand image content for creative modifications and also make fine local adjustments, especially skilled at handling text within images.

Some people from big companies talk about daily and monthly active users every day, but once they leave the big company, they don't know how to survive. 😂 Liu Xiaopai's writing today is truly exquisite, every word is a gem.

Luo Yonghao is entering the podcast industry... It seems like every time I do something, a few months later I can see him behind it...

AIpai supports music and video recognition features, allowing the generation of music videos based on the created music.

Regarding humanity's final moat, I found an answer: stories. This is an insight I gained after reading "The Warm Technology." The book provides a key perspective: The birth of human consciousness is marked by our beginning to understand the world through stories. We process chaotic experiences into causal narratives to handle the dimension of time and predict the future. It can be said that the ability to construct stories is the first core advantage that distinguishes humans from other species. A more central inference in the book is: Consciousness itself is the mechanism that transforms vast unconscious information into "stories." This model is very clear: the unconscious areas of the brain (driven by intuition, emotion, and desire) are the true decision-makers. Our perceivable consciousness plays the role of a spokesperson, tasked with rationalizing and narrating the unconscious decisions before outputting them. This mechanism explains the fundamental reason for the inconsistency between knowledge and action—the final action directive needs to receive approval from the unconscious. Behind your inability to resist scrolling through Douyin lies a humanity that even AGI cannot reach. Stories are the starting point of consciousness, and then stories elevate into a more advanced intelligence: empathy. When a person learns and categorizes enough stories, they gain the ability to understand others from their perspective. Empathy is the source of imagination and the cornerstone of trust. This made me realize where the boundaries that AI finds difficult to cross lie. AI can process knowledge, but it lacks a chaotic unconscious core that needs to be "narrated." Therefore, it cannot truly possess the profound empathy and trust generated by the layered progression of stories. Humanity's strongest moat is not our ability to process information, but our essence as natural "storytellers." As Einstein said: Imagination is more important than knowledge, for knowledge is limited, while imagination encompasses the entire world.

This infinite canvas AI is really too 🐮🍺 cool Just drop an image, sketch a few strokes, and you can get results Even the input prompt is saved This is the first project Jaaz from the Way2AGI community Jaaz is an open-source multimodal creative AI agent, pioneering a magical canvas for easy content creation.

Regarding humanity's final moat, I found an answer: stories. This is a realization I had after reading "The Warm Technology." The book provides a key perspective: the birth of human consciousness is marked by our beginning to understand the world through stories. We process chaotic experiences into causal narratives to handle the dimension of time and predict the future. One could say that the ability to construct stories is the first core advantage that distinguishes humans from other species. A more fundamental inference in the book is that consciousness itself is the mechanism that transforms vast unconscious information into "stories." This model is very clear: the unconscious areas of the brain (driven by intuition, emotion, and desire) are the true decision-makers. Our perceivable consciousness plays the role of a spokesperson, tasked with rationalizing and narrating the unconscious decisions before outputting them. This mechanism explains the fundamental reason for the inconsistency between knowledge and action—the final action directive needs to receive approval from the unconscious. Behind your inability to resist scrolling through Douyin lies a humanity that even AGI cannot reach. Stories are the starting point of consciousness, and then stories elevate into a more advanced intelligence: empathy. When a person learns and categorizes enough stories, they gain the ability to understand others from their perspective. Empathy is the source of imagination and the cornerstone of trust. This made me realize where the boundaries that AI struggles to cross lie. AI can process knowledge, but it lacks a chaotic unconscious core that needs to be "narrated." Therefore, it cannot truly possess the profound empathy and trust generated by the layered progression of stories. Humanity's strongest moat is not our ability to process information, but our essence as natural "storytellers." As Einstein said: Imagination is more important than knowledge, for knowledge is limited, whereas imagination encompasses the entire world.

The book written by the creator of Lovot is amazing; after reading it, I also want to create my own robot. The author's thoughts on robots and artificial intelligence are profoundly admirable. He is very clear about the boundaries of large models, knowing that even with AGI, robots cannot become human. This is also why large model companies cannot dominate all applications. Large models only simulate the conscious parts of humans, while the vast majority of human behavior is controlled unconsciously. For example, you know that scrolling through Douyin is wrong, but you just can't stop. For example, you know you shouldn't be nervous when giving a speech, but you still get so nervous that you can't speak. For example, you know that going to bed early and waking up early is good for your health, but you just can't help but stay up late. He also sees through the essence of robots and humans. Therefore, his judgment is that robots will absolutely not threaten humans. Because humans will die, and the selfish gene of humans aims to prolong life. Thus, humans often prioritize self-preservation. Because we cannot resist the instinct to prolong life and reproduce. We often make instinctive choices consciously and unconsciously (more often unconsciously). And although robots have intelligence, they do not have this selfish gene and do not have the need to prolong life. The fundamental reason for the existence of robots is altruism. Therefore, robots can become true partners of humans. These profound thoughts and insights make one reflect that the core of embodied intelligence may not be embodiment, nor intelligence, but rather the unconscious part beneath intelligence.

Yesterday I spoke at the CSDN Product Manager Conference. Three months ago, when my friends at CSDN invited me to give a talk at the Product Manager Conference, I actually wanted to decline. The reason was that my startup had only been established for six months, and I didn't have much valuable insight to share with everyone. However, my friends at CSDN said it was okay, the conference was still three months away, and I could also share some previous product experiences. As luck would have it, after we launched FlowSpeech last week, the product received rave reviews, our MRR tripled, and our ARR also surpassed a small target. More importantly, our users made real money by using the product, which is the best proof of our product's strength. So yesterday during my speech, I joked with everyone that the value of this PPT has skyrocketed, so please listen carefully.