ManyPI
Turn websites into APIs
490 followers
Turn websites into APIs
490 followers
Convert any website into a clean, type-safe API. Use natural-language or JSON-schema prompts to extract data and get usable JSON output for RAG pipelines, sales, content aggregation, or research.
Interactive
Free Options
Launch Team / Built With
Langswap.app - a video translator tool
Hi, thank you for the launch, I believe
is it possible to add some website for scrapping via API?
my case is mostly about one-time parsing with mostly static data. However if I got it correct current pricing tariffs are more about "per API". Is it possible to add tariff per API usage?
ManyPI
@nikolai_pakhtusov Hi Nikolai,
Yes, you can easily add any website you’d like to scrape via the API.
Our pricing is primarily usage-based. Each time a scraper runs—whether triggered via an API call or through our visual dashboard—the usage is tracked accordingly. In addition, each plan includes a defined number of active scrapers, depending on the selected tier.
If you anticipate needing significantly more active scrapers than what’s currently included in your plan, just send me a quick email. I’m happy to increase your active scraper limit free of charge. This is a soft limit that we’ve put in place and (if needed) can be expanded in certain cases.
I'm excited to welcome you as a ManyPI user. :)
Best,
Ole
Schema Generation Failed
Schema generation is temporarily busy. Please try again in a moment.
Is this service currently experiencing issues?
Or does this error occur on sites that don't support scraping?
ManyPI
@___83 Resolved this!
The error was caused by a temporary surge in demand, which led to a brief overload.
This was unrelated to any website-specific restrictions or scraping limitations. The system has since recovered automatically as capacity stabilized.
To prevent similar issues in the future, we have scaled up our infrastructure and implemented additional measures for automatic resource scaling during unusually high spikes in demand.
Please feel free to try again. :)
I love that it supports natural-language prompts makes data extraction feel way more approachable for non-developers too.
ManyPI
@rahim_pirzada Thank you! :)
ManyPI
@filblack Regarding use cases: During beta testing, we observed that researchers and PhD students primarily used our service to extract public data from publicly accessible sources (for example, government websites) that do not provide an API.
Additional use cases include the creation of datasets for AI fine-tuning, the implementation of RAG pipelines (more to come here soon), building internal databases, among others.
Regarding the risk levels: I agree that our current risk assessment is overly conservative. We did that to prevent unlawful usw. Yet we actively work on a fix, allowing for a wider range of sites to be scraped, and expect to release it soon.
The update will automatically detect and analyze the Terms of Service of the requested website. If the ToS does not explicitly prohibit scraping, we will lower the risk score, as scraping publicly available data is generally seen as permitted.
However, websites that explicitly restrict scraping and primarily contain personal data (for example, LinkedIn) will remain blocked/limited.
Nice idea! Is there any chance that Cloudflare will block the scraper?
ManyPI
@zakiah_hashim
So far, we’ve been able to avoid Cloudflare blocks. While a 100% success rate can’t be guaranteed, we’re not aware of any cases where our scraping engines have been blocked by target sites yet. This is thanks to the many sophisticated technical measures we’ve put in place to reliably prevent bot flagging.
@olemai cool. Thanks!
ManyPI
@zakiah_hashim Sure, you're welcome to try our service any time you like. :)
Tried it, but one run consumed all the credits and created nice structured output, but grabbed the wrong number. Another attempt to get an exchange rate was too risky.
ManyPI
@thaiguy Tokens are consumed based on the volume of data being processed. Some scrapers require manual approval when they relate to potentially high-risk areas, such as finance or the handling of personally identifiable data.
Please send a brief email to hello@manypi.com, and I’ll be happy to look into this for you.
ManyPI
@kandalia_kiki Each request has a token limit (system default), to protect users from unexpectedly high token consumption. We're also in the process of implementing custom token limits which can be set by users.
ManyPI
@kandalia_kiki This feature has now been implemented. You can set your custom limits in the account settings.
Agnes AI
Very interesting idea! What is the most popular scenarios for using ManyPI from your side?
ManyPI
@cruise_chen Academic research, internal database construction, competitor analysis, content planning, identification of SEO content gaps, creation of structured training data for fine-tuning AI models, and the implementation of RAG pipelines - a few scenarios where using ManyPI makes sense. :)