Top all-time · #1226
Jina AI
Company blog · jina.ai/newsaiprogramming
Jina AI is the company blog at jina.ai, writing about ai and programming. Since 2020, 58 different Hacker News readers have linked it 98 times, most often “Jina reader – convert webpage into Markdown for LLM”. No feed was found, so read it on the site itself.
Rarely recommended · 2 of 32 sampled comments recommend it?
Why HN links it
HN readers link Jina AI for its Reader API that turns web pages into Markdown, its embedding models, and its news posts on agentic search and late chunking, which the commenters cite as sources.
Summarised by Claude from a sample of 32 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Embeddings, rerankers, web reader, deepsearch, small language models. Search AI for multilingual and multimodal data.
Most-linked posts?
- Jina reader – convert webpage into Markdown for LLM8 people · 9 mentions · first linked 2024-04-16
- A Practical Guide to Implementing DeepSearch/DeepResearch6 people · 6 mentions · first linked 2025-02-27
- SEO Is Dead, Long Live LLMO5 people · 5 mentions · first linked 2022-12-15
- What Is ColBERT and Late Interaction and Why They Matter in Search?4 people · 4 mentions · first linked 2024-03-01
- Search Is Overfitted Create; Create Is Underfitted Search3 people · 4 mentions · first linked 2022-11-17
- Benchmarking Midjourney/Describe, CLIPinterrogator2.1, BLIP2 and SceneX2 people · 3 mentions · first linked 2023-04-04
- /2 people · 3 mentions · first linked 2020-06-02
- /2 people · 2 mentions · first linked 2020-12-26
- /news/late-chunking-in-long-context-embedding-models/2 people · 2 mentions · first linked 2024-09-20
- Why Don't People Use Your Software Framework?2 people · 2 mentions · first linked 2022-12-01
What Hacker News says?
TIL Jina Reader API has (low rate limit) API key free option. https://jina.ai/reader/
Comment linking /reader/
on “Everything I built with Claude Artifacts this week”Submitted Reader-LM: Small Language Models for Cleaning and Converting HTML to Markdown as a story · 199 points
HN discussionJina AI - https://jina.ai | Berlin/Beijing | Onsite Jina AI is a Neural Search Company, found in Feb. 2020 and closed the seed round in Mar. Valuation at $10M. We provide the cloud-native neural search solution powered by state-of-the-art AI and deep learning. We are hiring talents all over the world to build the next neural search eco-system in open-source. Checkout our core product "Jina": https://get.jina.ai Open…
Comment linking /
on “Ask HN: Who is hiring? (June 2020)”Submitted One Year in Open-Source Neural Search: Jina AI's 2020 Year in Review as a story · 12 points
HN discussionSurprised they never mentioned jina.ai models such as jina-embeddings-v3 at 8K context and outperforming most "contenders" in the MTEB or jina-clip-v2 (multimodal) or "late chunking," also known as mean pooling: https://jina.ai/news/late-chunking-in-long-context-embedding... The article feels incomplete
Comment linking /news/late-chunking-in-long-context-embedding-models/
on “Embeddings are underrated (2024)”Submitted Auto-GPT Unmasked: The Hype and Hard Truths of Its Production Pitfalls as a story · 7 points
HN discussionJina AI | https://jina.ai | Senior/Lead Python Engineer • Lead Dev Rel Specialist | Remote • Onsite (Berlin/Beijing) We're building "Tensorflow for search" in open-source. https://github.com/jina-ai/jina We are founded in 2020, backed by GGV Capital, working on Search AI ecosystem. We are expanding our R&D team and looking for engineers, dev-rels who are passionate about neural search, opensource and enjoy working i…
Comment linking /
on “Ask HN: Who is hiring? (February 2021)”Submitted Time-Slicing GPUs with Karpenter as a story · 5 points
HN discussionI haven't done exhaustive testing of all top-performing models on the HF Embedding Leaderboard ( https://huggingface.co/spaces/mteb/leaderboard ) but I have tested a number of them extensively in the past month or two. The two best API provider models I've tested are: - JinaAI ( https://jina.ai/embeddings/ ) v3 and v4 performed well in my testing. - Google's Gemini-001 model ( https://ai.google.dev/gemini-api/docs/m…
Comment linking /embeddings/
on “LLM Embeddings Explained: A Visual and Intuitive Guide”Submitted Speech-to-image creation using Jina as a story · 5 points
HN discussionYes! Han Xiao at Jina wrote a great article that goes into a lot more detail on how to turn this into a production quality agentic search: https://jina.ai/news/a-practical-guide-to-implementing-deeps... This is the same principle that we use at Brokk for Search and for Architect. ( https://brokk.ai/ ) The biggest caveat: some models just suck at tool calling, even "smart" models like o3. I only really recommend Gemi…
Comment linking /news/a-practical-guide-to-implementing-deepsearch-deepresearch/
on “The unreasonable effectiveness of an LLM agent loop with tool use”Submitted Jina-OCR-v1: Faster Document Parsing on Low-Budget GPU as a story · 4 points
HN discussionThe recent project from jina AI was a clone of this idea? https://jina.ai/reader/
Comment linking /reader/
on “Show HN: I made a tool to clean and convert any webpage to Markdown”Submitted A Practical Guide to Implementing DeepSearch/DeepResearch as a story · 4 points
HN discussionWe're a cutting edge AI company doing neural search, seeking AI engineers and full-stack developers who actually give a crap about open source. A lot of people talk a big game about how awesome open-source software is. Not so many really understand its value and believe in it so much that they'd pack in their stable jobs at tech giants and risk it all for an open-source venture. We're in that second group. A lot of…
Comment linking /
on “Ask HN: Who is hiring? (July 2020)”Submitted Hype and Hybrids: Search is more than Keywords and Vectors as a story · 4 points
HN discussionIf you need this sort of thing in any other language, there's a free, no-auth, no-api-key-required, no-strings-attached API that can do this at https://jina.ai/reader/ You just fetch a URL like ` https://r.jina.ai/https://www.asimov.press/p/mitochondria `, and get a markdown document for the "inner" URL. I've actually used this and it's not perfect, there are websites (mostly those behind Cloudflare and other such p…
Comment linking /reader/
on “Show HN: HTML-to-Markdown – convert entire websites to Markdown with Golang/CLI”Submitted ReaderLM v2: Frontier Small Language Model for HTML to Markdown and JSON as a story · 3 points
HN discussionhttps://www.xaynet.dev/ - open source federated learning backend in Rust https://flower.dev/ - A Friendly Federated Learning Framework in Python https://jina.ai/ - Neural search engine
Comment linking /
on “Ask HN: What startup/technology is on your 'to watch' list?”Submitted Jina Embeddings V3: A Frontier Multilingual Embedding Model as a story · 3 points
HN discussionAnother example. I have web pages fetching. But web page fetching bring a lot of JS / noise. So the fetched page instead use https://jina.ai/reader/ . I get Markdown. But is this enough. No, there still a lot of links and stuff, so I again do another pipe and pass to strip a lot like URL's I usually don't need as focus here on content.
Comment linking /reader/
on “LLM function calls don't scale; code orchestration is simpler, more effective”Submitted What Is ColBERT and Late Interaction and Why They Matter in Search? as a story · 3 points
HN discussionJina did something related for extracting content from raw HTML and wrote about the techniques they used here: https://jina.ai/news/reader-lm-small-language-models-for-cle... .. in my tests, the 1.5bn model works extremely well, though the open model is non commercial.
Comment linking /news/reader-lm-small-language-models-for-cleaning-and-converting-html-to-markdown/
on “Extracting financial disclosure and police reports with OpenAI Structured Output”Submitted Jina reader – convert webpage into Markdown for LLM as a story · 3 points
HN discussionI really like the distinction between DeepSearch and DeepResearch proposed in this piece by Han Xiao: https://jina.ai/news/a-practical-guide-to-implementing-deeps... > DeepSearch runs through an iterative loop of searching, reading, and reasoning until it finds the optimal answer. [...] > DeepResearch builds upon DeepSearch by adding a structured framework for generating long research reports Given these definitions…
Comment linking /news/a-practical-guide-to-implementing-deepsearch-deepresearch/
on “The Differences Between Deep Research, Deep Research, and Deep Research”Submitted SEO is dead long live LLMO as a story · 3 points
HN discussionFor a Regex approach take a look at the work from Jina.ai who among other things have a chunk/tokenizer [1] and now it's part of a bigger API service [2] also they developed an interesting late interaction (aka ColBERT like) chunking system that fits certain use cases. But the Regex is enough all by itself: [1] https://gist.github.com/LukasKriesch/e75a0132e93ca989f8870c4... [2] https://jina.ai/segmenter/
Comment linking /segmenter/
on “Show HN: Chonkie – A Fast, Lightweight Text Chunking Library for RAG”Submitted SEO Is Dead, Long Live LLMO as a story · 3 points
HN discussionfor web PDFs I'm using https://jina.ai/reader/ — completely free. Does most of the job fine. Code: https://github.com/jina-ai/reader
Comment linking /reader/
on “Ask HN: What are you using to parse PDFs for RAG?”Submitted Why Don't People Use Your Software Framework? as a story · 3 points
HN discussion
