Gemini

A large model that can quickly analyze videos; its recognition capability is very strong in my personal testing. Address: Upload video Prompt: Please deconstruct this video, mainly analyzing the detailed footage of the production process and the spoken script, and tell me the author's design approach. After the author selects the theme of this video, how does he design the script, including the characteristics of his copy? If the copy is broken down, it can be analyzed separately from small modules of content structure, content elements, technique analysis, emotional manipulation chain, etc., and finally give me a reusable formula. Read more

At the same time Sora was released, Google released Gemini 1.5. But compared with Sora’s stunning capabilities, Gemini seems far too underwhelming. After this 1.5 release, we can communicate in Chinese, and image generation is supported. Unfortunately, image ... Read more

Gemini has released 1.5. It now supports Chinese input and can also generate images, but the image generation functionality all uses other platforms. Also, note that for image generation you must use English. If you use Chinese input, it will prompt you that it cannot create. When I asked it what its image generation function was, it answered me that it did not have one. Read more

First, we asked it whether it could generate comic images, and then it told me it could. What I didn't expect was that it actually mentioned many other tools—could it be that it doesn't have drawing capabilities? Next, I asked it what its drawing function was... Read more

Gemini official address; as of the publication time of this article, it currently supports Chinese input and text-to-image features. Webpage address: The homepage displays its main functions, such as text-to-text, text-to-image, translation, linking to YouTube... Read more

https://www.bilibili.com/video/BV152421A7hS/?spm_id_from=333.337.search-card.all.click Brief Illustrated Guide GEMINI, which is comparable to GPT-4 and free to use for 60 days, has been released and is very suitable for beginners to try Image generation Image analysis; it can analyze math, jokes, logic, and more Articles with both text and images are no problem at all Read more

The text-to-image feature is now available and supports Chinese input. The Gemini text-to-image feature has not yet been fully opened, but based on past GPT DALL-E 3 models, it will generally follow the methods below. Please stay tuned; this site will publish ... Read more

Gemini is the first AI model ever launched by Google with native multimodal capabilities, and it has been used in Google Pixel 8 Pro smartphones and the chatbot Bard. Currently, text-to-image is not yet supported, but it won't be long; the technology has alrea... Read more

The registration and usage tutorial for Google Gemini is as follows (note: operate under a VPN): The address is as follows: Click the blue button Then log in to your Google account Interface display If all goes well, you will enter Black is the default page; c... Read more

https://www.bilibili.com/video/BV1b64y157Xt/?spm_id_from=333.337.search-card.all.click&vd_source=383e491c6ed7635b9a5413557bd39010 Many people say Gemini crushes GPT—is that really true? Accuracy comparison between the Gemini model and the GPT-4 model: 90:86.4 Gemini itself says it only performs well under the Cot@32 model Read more