43
0Episode: DeepSeek V4 Flash Vision Exp: AI Learns to See
Duration: approximately 7 minutes
Level: B1 (Intermediate)
---
[Mike]: Welcome back to Learn English with Podcasts! Sarah, quick question - can your AI read your vacation photos and actually do something useful with them?
zh:欢迎回到 Learn English with Podcasts！Sarah，问你个小问题——你的 AI 能看懂你的度假照片，然后真的做点有用的事吗？
[Sarah]: Honestly, most AIs just say nice photo! But you're smiling like you found one that does more.
zh:老实说，大多数 AI 只会说"照片不错！"但你笑得这么开心，肯定是发现了更厉害的。
[Mike]: I did. DeepSeek just released something called DeepSeek V4 Flash Vision Exp. It's experimental - Exp means it's a test version.
zh:我确实发现了。DeepSeek 刚刚发布了一个叫 DeepSeek V4 Flash Vision Exp 的东西。它是实验性的——Exp 就是测试版的意思。
[Sarah]: DeepSeek V4 Flash Vision Exp - that's a long name! What makes it special?
zh:DeepSeek V4 Flash Vision Exp——名字好长啊！它特别在哪里？
[Mike]: Two words: it can see. The normal V4 Flash is great with text. This new one keeps all that text power, but adds eyes.
zh:两个字：它能看见。普通的 V4 Flash 就已经很擅长处理文字了。这个新版本保留了全部文字能力，还加了一双眼睛。
[Sarah]: Wait, so it didn't lose any text skills? Usually when you add vision, the text part gets weaker.
zh:等等，所以它的文字能力一点没丢？通常加了视觉能力后，文字部分就会变弱。
[Mike]: Exactly, that's the balance. In pure text tasks - like solving problems or answering knowledge questions - it scores the same as V4 Flash.
zh:没错，这就是它的平衡。在纯文字任务上——比如解决问题或回答知识类问题——它的得分和 V4 Flash 一样。
[Sarah]: Okay, that's good. But what about the vision part? How much better is it?
zh:好，这很不错。那视觉部分呢？提升了多少？
[Mike]: Big jump. On agent benchmarks that need vision, the normal Flash was not strong. This Vision Exp version jumps up to almost the level of Opus-4.8.
zh:提升很大。在那些需要视觉的智能体测评里，普通的 Flash 并不强。这个 Vision Exp 版本一跃提升到几乎和 Opus-4.8 同等水平。
[Sarah]: Opus-4.8? That's one of the top models. So this experimental model is already playing in the big league?
zh:Opus-4.8？那可是顶级模型之一。所以这个实验性模型已经能和顶尖选手同场竞技了？
[Mike]: Yes. And the cool part is how it works inside agent tools. You give it a big goal, and it uses vision to get there.
zh:是的。更酷的是它在智能体工具里的工作方式。你给它一个大目标，它就用视觉能力去实现。
[Sarah]: Give me a real example. I learn better with stories.
zh:给我一个真实的例子。我听故事学得更快。
[Mike]: Okay, first one - a luxury Tibet travel PPT. The prompt was crazy detailed: one month, self-driving, south plus north Tibet, wild and raw style, not touristy photos.
zh:好，第一个——一份高端西藏旅行 PPT。提示词非常详细：一个月，自驾，藏南加藏北，要野性原始的风格，不要游客照。
[Sarah]: A month in Tibet? And the AI has to make slides for rich customers?
zh:在西藏待一个月？而且 AI 要给高净值客户做幻灯片？
[Mike]: Yes, for high-end private tours. And get this - the final PPT needed three real pricing plans, with real photo style images, rough and powerful look. No cute filters.
zh:对，是给高端私人定制游的。而且关键是——最后的 PPT 需要三种真实的报价方案，配上真实摄影风格的图片，粗粝有力量感。不要小清新的滤镜。
[Sarah]: And the AI did it? Just from that long text description?
zh:然后 AI 做出来了？就凭那段长长的文字描述？
[Mike]: With the vision ability, yes. It found the right images, matched the wild tone, and built a real-looking deck that you could actually send to customers.
zh:有了视觉能力，是的。它找到了合适的图片，匹配了野性的基调，做出了一份看起来真的能发给客户的演示文稿。
[Sarah]: Wow, so it understands both the words and the pictures fit the feeling. What's the second example?
zh:哇，所以它既理解文字，也能让图片贴合感觉。第二个例子是什么？
[Mike]: This one is fun. They asked it to rebuild a developer website - the DeepSeek Harness site - with a future style. Dark blue sea, glass UI, ASCII art.
zh:这个很有趣。他们让它重做一个开发者网站——DeepSeek Harness 的官网——要未来感风格。深蓝色海洋、玻璃质感界面、ASCII 艺术字。
[Sarah]: ASCII art? Like old computer letters making pictures?
zh:ASCII 艺术？就是用老式电脑字母拼成图片那种？
[Mike]: Exactly. And after many rounds of talking, the model turned that idea into a whole new site. It kept listening, kept seeing what it made, and fixed it step by step.
zh:没错。而且经过多轮对话后，模型把那个想法变成了整个新网站。它一边听，一边看自己做出来的东西，然后一步步修改。
[Sarah]: So it's not one-shot. It's like working with a designer who shows you drafts?
zh:所以不是一次成型。就像和一个会给你看草稿的设计师合作？
[Mike]: Perfect example. And third one - a mini website with cute 3D clay monsters dancing at a retro party.
zh:完美的比喻。第三个——一个迷你网站，一群可爱的 3D 黏土小怪物在复古舞池里跳舞。
[Sarah]: Clay monsters dancing? That's so random. I love it.
zh:黏土怪物跳舞？这也太随意了吧。我喜欢。
[Mike]: The prompt was random too: a group of cute 3D clay monsters joining a jumping retro dance party. And the agent built a moving demo in the browser.
zh:提示词本身就很随意：一群可爱的 3D 黏土小怪物加入一个跃动的复古舞池派对。然后智能体直接在浏览器里做出了会动的演示。
[Sarah]: Okay, so it can see, understand tone, and build things. How do developers actually use it? Is it hard?
zh:好，所以它能看，能理解基调，还能搭建东西。开发者实际怎么用它？难吗？
[Mike]: Super easy. You just set model='deepseek-v4-flash-vision-exp' in the API. That's it.
zh:超级简单。你只需要在 API 里设置 model='deepseek-v4-flash-vision-exp' 就行。就这么简单。
[Sarah]: Just change one line? And the price?
zh:就改一行？那价格呢？
[Mike]: Same as V4 Flash. Each image costs at most 384 tokens. So you pay by tokens, not by extra vision fee.
zh:和 V4 Flash 一样。每张图片最多算 384 tokens。所以你按 token 付费，没有额外的视觉费用。
[Sarah]: 384 tokens per image - that's pretty cheap. What about sending images?
zh:每张图片 384 tokens——挺便宜的。那怎么发送图片呢？
[Mike]: Three ways: you can paste base64 code, send a URL link, or use their new Files API.
zh:三种方式：你可以粘贴 base64 编码，发一个网址链接，或者用他们新的 Files API。
[Sarah]: Files API? Is that new?
zh:Files API？是新出的吗？
[Mike]: Yes, just launched. You upload the image once, get a file_id, and use that file_id in many requests. No need to upload again. And it's free to upload.
zh:是的，刚刚上线。你上传一次图片，拿到一个 file_id，然后在很多请求里重复使用这个 file_id。不用重复上传。而且上传是免费的。
[Sarah]: Oh, that saves time and bandwidth. Does the API work with different formats?
zh:哦，那省了时间和带宽。API 支持不同的格式吗？
[Mike]: Yes, three formats: Chat Completions, Messages, and Responses. So it fits into most agent tools easily.
zh:支持，三种格式：Chat Completions、Messages 和 Responses。所以它能很轻松地接入大多数智能体工具。
[Sarah]: You know what I find interesting? This is an experimental model. Exp means they are still testing.
zh:你知道我觉得有意思的是什么吗？这还是个实验性模型。Exp 意味着他们还在测试。
[Mike]: Right, but it's already strong enough to make real PPTs and websites. Imagine what the final version will do.
zh:对，但它已经强到能做真正的 PPT 和网站了。想想最终版会做到什么程度。
[Sarah]: And it mixes text and images together. That feels more human. We don't just read - we look.
zh:而且它把文字和图片混在一起处理。这感觉更像人类。我们不只是阅读——我们也在看。
[Mike]: Right. Before, AI was like a smart friend who could only hear your words. Now it can see your photos, your screenshots, your designs.
zh:对。以前，AI 就像一个只能听你说话的聪明朋友。现在它能看到你的照片、截图、设计稿。
[Sarah]: So it's not about who talks the most. It's about who can see what you're pointing at.
zh:所以重要的不是谁话最多，而是谁能看懂你在指什么。
[Mike]: Exactly! And here's a funny thought - we spent years teaching AI to read our messy handwriting. Now we're teaching it to judge our PowerPoint taste.
zh:没错！还有个好笑的想法——我们花了好多年教 AI 认我们潦草的字迹。现在我们在教它评判我们的 PPT 审美。
[Sarah]: Ha! My PPT taste definitely needs judging. So if listeners want to try it, just try that model name?
zh:哈！我的 PPT 审美确实需要被评判一下。所以如果听众想试试，只要试试那个模型名就行？
[Mike]: Yes, try model DeepSeek V4 Flash Vision Exp and see what it sees. And tell us - what would you ask an AI that can finally see?
zh:对，试试模型 DeepSeek V4 Flash Vision Exp，看看它能看见什么。也告诉我们——如果 AI 终于能看见了，你会让它帮你做什么？
[Sarah]: We'd love to hear your ideas. See you next time on Learn English with Podcasts!
zh:我们很期待听到你的想法。下期 Learn English with Podcasts 再见！
[Mike]: See you next time!
zh:下期见！
