AI News

DeepSeek packages web search, document parsing, and voice capabilities as open APIs

AYi AI NotesUpdated Nov 14 20:59

AI summary

DeepSeek has packaged the web search, PDF and Office document parsing, web link scraping, security screening, as well as OCR, ASR, and TTS capabilities from its official App into open APIs. Currently, these are only available for deepseek-flash (V4.1-Flash) and are enabled via parameter switches when calling. On pricing: web search and link scraping cost $0.75 per 1,000 calls, document parsing costs $0.225 per 1,000 pages, speech recognition costs $0.058 per hour, and off-peak input tokens cost $0.15 per million. Meanwhile, DeepSeek has open-sourced the chat-demo frontend project on GitHub under the MIT license; you can run it locally by double-clicking index.html, and it can import chat history exported from the App.

Why read this

AI score 75

The original article presents DeepSeek's newly opened capabilities—search, document parsing, voice, and more—alongside specific call pricing, making it easy for developers to estimate the cost of building their own applications.

Topics

Tags

Quoted post

AYi

@AYi_AInotes

Fully open-sourced!! I used Tencent WorkBuddy to run a complete Hong Kong and US stock AI investment research workflow: reviewing holdings, breaking down financial reports, and relentlessly tackling data hallucinations, with a 4-step prompt included—I suggest copying it directly. As long as you regularly read financial reports and manage holdings, you can directly use this human-AI collaboration as an AI investment research SOP workflow for future investment research reviews: the core logic is not to have AI predict tomorrow's gains or losses like a fortune teller, but to treat it as a cold auditor that thoroughly verifies the chain of evidence behind every number. Perhaps only after calculating this cost will everyone understand how important this system is: What you fear most when reading financial reports is never misjudging the major trend, but rather AI fabricating fake data, or mixing in accounting standards restated by the company in the following year, and your real-money decisions worth tens of thousands or even hundreds of thousands instantly hit a landmine; This workflow illuminates all the hidden pitfalls, and even if you lose money, you can review it clearly! I have always believed that what AI should really do in investment research is the most labor-intensive, dirty, and tedious work. I summarize it as conducting a background check on numbers, every piece of data must pass three checks: → Where it came from: which section of the annual report, and clicking it must trace back to the original text → What it is compared with: whether the accounting standards across periods are the same, and whether they can be directly subtracted → Who said it: whether it is a verified fact or the management's own explanation. Numbers whose background cannot be clearly checked are not allowed into the decision conclusions. The video contains the complete screen recording (all demonstrations use anonymized synthetic data), and the entire test is divided into four extremely clean steps: Build a holdings dashboard; if cost is not provided, resolutely do not guess. Drop in the raw holdings and price snapshot, and convert weights into Hong Kong dollars by currency. For positions without recorded cost, honestly mark the cost as missing, and prevent model hallucination interpolation. Dynamic concentration risk control. After adding buy records, automatically recalculate market value. When a single holding is diluted to 29.18% due to adding to the position, the concentration warning is automatically withdrawn, and the before and after of the rebalancing are clear at a glance. Investigate the accounting standard traps of annual report restatements. Break down business segment growth, and when the company restates standards across periods, numbers are precisely anchored to the original annual report section, and clicking it shows the restatement notes, avoiding misleading fake increments. Ultimate follow-up question, three-column physical isolation. When encountering abnormal accounts receivable, force it to break down into three columns: verified facts, the company's own explanations, and judgments still requiring manual verification, completely isolating management's empty promises from objective facts. Finally, I want to tell everyone that investment research tools are never crystal balls; what is more terrifying than losing money is losing according to no system of your own. Let's encourage each other! The original text of the 4-step prompt and experience guide are organized in the comments section; as usual, help yourselves👇

View original

全盘开源!!我用腾讯 WorkBuddy 跑通了一套港美股 AI 投研工作流:复查持仓、拆解财报、死磕数据幻觉,附 4 步提示词,建议直接照着抄。 只要你平时看财报、管持仓, 以后做投研复盘可以直接拿这套人机协作当AI投研sop工作流:核心逻辑不是去让 AI 算命猜明天涨跌,而是把它当成冷酷的审计员,把每一笔数字背后的证据链彻底查清。 可能算过这笔账大家才知道这套体系有多重要: 平时看财报最怕的从来不是看错大趋势, 而是 AI 瞎编一个假数据, 或者把公司第二年重列的会计口径混着算, 你几万甚至几十万的实盘决策就当场踩雷; 而这套工作流把所有暗坑全部照亮, 亏了钱也能清清楚楚复盘! 我一直觉得AI 在投研里真正该干的就是最费人工的脏活累活。 我把它总结为给数字查户口, 每一笔数据都必须过三关: → 哪来的:出自年报哪一节,点开必须能追溯到原文 → 跟谁比:跨期会计口径一不一样,能不能直接相减 → 谁说的:是已经核实的事实,还是管理层自己的解释 户口查不清的数字,一个都不准进决策结论。 视频里是完整录屏(演示全部使用匿名合成数据), 整套实测分为极度干净的四步: 做持仓看板,没给成本坚决不瞎猜 丢入原始持仓和价格快照,按币种折算港币权重。遇到未录入成本的头寸,老老实实标出成本缺失,杜绝模型幻觉插值。 集中度动态风控 补录买入记录后自动重新折算市值。 单只持仓因加仓被稀释到 29.18% 时, 集中度预警自动撤销,调仓前后一目了然。 排查年报重列的口径陷阱 拆解业务分部增长, 遇到公司跨期重列口径时, 数字精准锚定在原始年报章节, 点开就能看到重列附注,避免假增量误导。 终极追问,三栏物理隔离 遇到应收账款异动, 强制要求它分三栏拆解: 已验证事实、公司自己的解释、还需人工核实的判断,把管理层画饼与客观事实彻底隔绝。 最后想跟大家说,投研工具从来不是水晶球,比亏钱更可怕的是没有按照自己的体系去亏,共勉! 4 步提示词原文与体验指引整理在评论区了,老规矩大家自取👇

View original post

Body · AI translation

Holy crap, everyone stand up!! Liang the Saint is a legend!! This time DeepSeek really pulled out every last trick it had hidden up its sleeve, packaging the whole set of capabilities from its official App—web search, PDF document reading, webpage link fetching, security checks, plus speech recognition and synthesis—and opening them all up as an API, and it even open-sourced the frontend interface directly under the MIT license.

A lot of you just getting started may not have realized how explosive this is, so let me break down the math for you in plain terms and you'll get it.

In the past, if you wanted to hand-build a decent AI application, the model itself could only chat; if you wanted it to search the web for information, you had to go buy a search engine API yourself,

if you wanted it to read contracts and PDFs, you had to write code yourself to parse the documents and then cram them into the context, and to add voice interaction,

you'd have to hook up yet another set of recognition and text-to-speech APIs—just writing all that glue code could torture half the life out of you.

Now DeepSeek has taken all this dirty, tedious work and done it all on its own server side.

As long as you call deepseek-flash (that is, V4.1-Flash) and add a few parameter switches to your request,

it can think while doing deep web searches, directly swallow PDF and Office files, automatically read webpage links in messages,

and it even comes fully equipped with input/output security screening and standalone OCR, ASR, and TTS.

Put simply, in the past, building an AI application meant going to the market yourself to buy groceries, going home to wash and chop the vegetables, and then firing up the stove to cook; now DeepSeek has hooked up the vegetable chopper, the oven, and the dishwasher for you—all you have to do is order the dishes and serve them.

Even more absurd is the price: web search and link fetching cost just $0.75 per thousand calls, parsing a thousand pages of documents costs just $0.225,

speech recognition costs just $0.058 per hour, and combined with the off-peak price of $0.15 per million input tokens, individual developers can get started at almost zero threshold, freeloader-style.

Interestingly, they also casually open-sourced a frontend project on GitHub called chat-demo—no need to set up a complicated environment,

just double-click index.html locally and it runs, and you can even directly import the chat history you exported from the DeepSeek App and seamlessly continue the conversation.

I've always felt that for ordinary people using AI to build side hustles or run one-person companies, the biggest fear is spending all your time reinventing the wheel,

although right now this full-power toolchain is only opened up for Flash first, and with high-frequency calls you also need to watch the number of searches so your bill doesn't blow up,

but looking at this trend, the technical barrier to hand-building AI products really has been flattened to the point where it's like stacking building blocks.

Source

AYi AI Notes