{"slug": "exa", "name": "Exa", "category": "search", "category_name": "Search", "description": "", "recommenders": 10, "commenters": 24, "warners": 3, "score": 9.192, "rising": 3.36, "first": "2025-12-22", "latest": "2026-10-18", "why": "try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 10, "neutral": 5, "compare": 6, "warn": 3}, "spark": [0, 0, 2, 0, 0, 0, 1, 0, 2, 3, 1, 1], "listed": true, "loved": [], "caveat": "", "short_quote": "try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets.", "kind": "", "icon": true, "domain": "exa.ai", "github_repo": null, "official_url": "https://exa.ai", "pricing": "", "aliases": ["Exa", "exa"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 802235840, "item_id": 50139740, "url": "https://news.ycombinator.com/item?id=50139740", "by": "vitorsr", "date": "2026-10-18", "version": null, "stance": "recommend", "quote": "I pay the Exa premium for semantic retrieval with far higher SNR, plus the academic index", "excerpt": "Meant as a comparison given the discrepant pricing points: I pay the Exa premium for semantic retrieval with far higher SNR, plus the academic index [1]. A field like objective can be up to 4096 characters [2] versus ori_question which is limited to 128 here. [1] https://exa.ai/blog/publications-search [2] https://exa.ai/docs/reference/search", "context": "", "use_case": "semantic search API", "switched_from": "", "story": "DeepSeek search and fetch \u2013 $0.75 per 1000 successful requests"}, {"id": 792424352, "item_id": 49526522, "url": "https://news.ycombinator.com/item?id=49526522", "by": "verdverm", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets.", "excerpt": "My experience is otherwise, try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets. My agent looks at the results, includes some, performs more searches. It's more of a deep search thing, very similar in spirit to how I used to run multiple queries, click a bunch of links, evaluate, keep good ones, run follow up queries, and end up with an alt-tab of dozens of b", "context": "", "use_case": "agent web search", "switched_from": "", "story": "Thanks to Lake Ontario, MapQuest is popular all over again"}, {"id": 791310176, "item_id": 49456886, "url": "https://news.ycombinator.com/item?id=49456886", "by": "usef-", "date": "2026-08-26", "version": null, "stance": "recommend", "quote": "Exa also has an API for it that has worked well for me", "excerpt": "Exa also has an API for it that has worked well for me, returning markdown for a URL, which means you don't need to render js or anything yourself. It doesn't need an account for up to 1k requests/month, which is more than I've ever needed. Most harnesses have a \"zero config\" plugin that works, eg: https://pi.dev/packages/pi-exa (no affiliation to either)", "context": "", "use_case": "Fetching URLs as markdown", "switched_from": "", "story": "Serve Markdown to AI Agents with Accept Headers"}, {"id": 789621008, "item_id": 49351313, "url": "https://news.ycombinator.com/item?id=49351313", "by": "fwn", "date": "2026-08-18", "version": null, "stance": "recommend", "quote": "exa.ai/about looks interesting to me", "excerpt": "Mozilla PR department seems to be vanilla Claude. tl;dr of the article: Rather than building a closed AI stack (like Google - but they don't name it) Mozilla is pursuing partnerships. In this case, they are partnering with Exa.ai to offer AI integrations on desktop and iOS devices with visible source citations. There is no mention of confidential computing, hardware enclaves, or other technical mechanisms but they pr", "context": "", "use_case": "AI search API", "switched_from": "", "story": "Firefox and Exa Partnership"}, {"id": 787829088, "item_id": 49239318, "url": "https://news.ycombinator.com/item?id=49239318", "by": "ignoramous", "date": "2026-08-09", "version": null, "stance": "recommend", "quote": "see exa.ai who build generic solutions in similar space", "excerpt": "It is both a great (ai tools) & worst (ai slop) time to be building a web search engine! > working on some more commercial features, like the ability to monitor queries (e.g. for backlinks or brand mentions) and get an email when there are new results As an inspiration (presuming you don't already know), see exa.ai who build generic solutions in similar space: https://exa.ai/docs/reference/monitors-api-guide", "context": "", "use_case": "Search API", "switched_from": "", "story": "Ask HN: What are you working on? (August 2026)"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 760239248, "item_id": 47514953, "url": "https://news.ycombinator.com/item?id=47514953", "by": "arthurjean", "date": "2026-03-25", "version": null, "stance": "warn", "quote": "Exa's MCP server is stateless, just a REST wrapper", "excerpt": "You're right actually. Exa's MCP server is stateless, just a REST wrapper. A skill + CLI would do the same job with way less context cost. Someone already built that (https://github.com/tobalsan/exa).", "context": "", "use_case": "", "switched_from": "", "story": "Welcome to FastMCP"}, {"id": 752056704, "item_id": 47003544, "url": "https://news.ycombinator.com/item?id=47003544", "by": "tibbar", "date": "2026-02-13", "version": null, "stance": "warn", "quote": "In practice, it's never worked that well for me", "excerpt": "Oh, like, I wouldn't actually use this specific implementation. I used to work at a shop with thousands of lines of Oracle triggers that you had to edit inline in the web browser with no version control and I shudder to think of returning again. I'm more interested in the data flow. exa.ai got famous for promising search with massively parallel execution of LLMs on candidate results. In practice, it's never worked th", "context": "", "use_case": "AI web search", "switched_from": "", "story": null}, {"id": 750242417, "item_id": 46890151, "url": "https://news.ycombinator.com/item?id=46890151", "by": "mheavers", "date": "2026-02-04", "version": null, "stance": "warn", "quote": "We switched from Exa because Tabstack proved more reliable at retrieving content from social media urls", "excerpt": "My team's product uses Tabstack's other structured data endpoints to retrieve the content of URLs for filling out of profile form fields and it has been flawless. We switched from Exa because Tabstack proved more reliable at retrieving content from social media urls", "context": "team building a product", "use_case": "retrieving URL content", "switched_from": "", "story": "Show HN: Tabstack Research \u2013 An API for verified web research (by Mozilla)"}]}], "who": [], "tags": [], "use_cases": ["semantic search api", "search api for llms", "agent web search", "fetching urls as markdown", "ai search api", "people search"], "compared": [{"slug": "searxng", "name": "SearXNG", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "claude", "name": "Claude", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "perplexity", "name": "Perplexity", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "claude-code", "name": "Claude Code", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "brave-search-api", "name": "Brave Search API", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [{"id": 799527296, "item_id": 49970456, "url": "https://news.ycombinator.com/item?id=49970456", "by": "alexey-salmin", "date": "2026-10-05", "version": null, "stance": "recommend", "quote": "We do have that in Exa search", "excerpt": "We do have that in Exa search: * Extraction of main content from HTMLs, the index stores markdown representations (free of headers/footers/sidebars/menus etc) * Serving highlights picking the most relevant part for each result to reduce token usage downstream * Also serving dynamic highlights where we summarize all the sources at once reducing the token counts even further see here: https://exa.ai/docs/search/highlig", "context": "", "use_case": "web search API, content extraction", "switched_from": "", "story": "Web Search API"}, {"id": 799524928, "item_id": 49970308, "url": "https://news.ycombinator.com/item?id=49970308", "by": "alexey-salmin", "date": "2026-10-05", "version": null, "stance": "recommend", "quote": "we perform well on all the verticals we optimize for", "excerpt": ">Here we are, one layer of indirection more: Ceramic, Exa, Linkup. Who knows what they use. If you told me that those 3 build and maintain their own index, I'd first question whether that was true, and if it is, I would question whether it was any good (relative to Google/Bing/Brave). Hi! Exa Head of Index here. We certainly do have our own index and it's one of the biggest among the independent players (i.e. not Goo", "context": "Head of Index at Exa", "use_case": "Web search API for AI", "switched_from": "", "story": "Web Search API"}], "threads": [{"url": "https://news.ycombinator.com/item?id=49963171", "title": "Web Search API", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=50138531", "title": "DeepSeek search and fetch \u2013 $0.75 per 1000 successful requests", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49435555", "title": "Show HN: Keenable \u2013 A different web search API for AI agents", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48835111", "title": "Grok 4.5", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=48266051", "title": "Search engines alternatives now that Google isn't Google anymore", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=49537553", "title": "Gemini 3.8 Flash and 3.8 Flash Cyber", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {"neutral": 1}, "2026-01": {"recommend": 2}, "2026-02": {"warn": 2}, "2026-03": {"neutral": 2, "warn": 1}, "2026-04": {}, "2026-05": {"compare": 1, "recommend": 1}, "2026-06": {}, "2026-07": {"recommend": 2}, "2026-08": {"neutral": 1, "recommend": 3, "compare": 3}, "2026-09": {"recommend": 1, "neutral": 1}, "2026-10": {"compare": 2, "recommend": 1}}, "versions": [], "version_unnamed": 0, "family_ref": null, "runs_on": null, "lines": [], "quotes": [{"id": 802235840, "item_id": 50139740, "url": "https://news.ycombinator.com/item?id=50139740", "by": "vitorsr", "date": "2026-10-18", "version": null, "stance": "recommend", "quote": "I pay the Exa premium for semantic retrieval with far higher SNR, plus the academic index", "excerpt": "Meant as a comparison given the discrepant pricing points: I pay the Exa premium for semantic retrieval with far higher SNR, plus the academic index [1]. A field like objective can be up to 4096 characters [2] versus ori_question which is limited to 128 here. [1] https://exa.ai/blog/publications-search [2] https://exa.ai/docs/reference/search", "context": "", "use_case": "semantic search API", "switched_from": "", "story": "DeepSeek search and fetch \u2013 $0.75 per 1000 successful requests"}, {"id": 802230787, "item_id": 50139424, "url": "https://news.ycombinator.com/item?id=50139424", "by": "wmchen", "date": "2026-10-18", "version": null, "stance": "compare", "quote": "Exa Fast / Auto: 2,233 searches/month", "excerpt": "Thanks for sharing, wasn't aware of this provider. If anyone is wondering about rate limits for the subscription: >We calculate search rates based on the timestamps of the last searches of a user (just the timestamps, not the queries or anything else!). We keep those timestamps for 12 hours, then discard them. We currently limit searches to 5 per 5 seconds, 15 per minute, and 120 per hour. Source: https://uruky.com/f", "context": "", "use_case": "", "switched_from": "", "story": "DeepSeek search and fetch \u2013 $0.75 per 1000 successful requests"}, {"id": 799441587, "item_id": 49965099, "url": "https://news.ycombinator.com/item?id=49965099", "by": "karmakaze", "date": "2026-10-05", "version": null, "stance": "compare", "quote": "Exa:        $7.00 per 1,000 requests", "excerpt": "I was just looking into these as DeepSeek Harness w/ Qwen3.8-27B relies heavily on search. I was going to go with Serper.dev[0] $1 per 1000 (or lower in quantity). The providers[1] behind this Web Search API have very different rates: Ceramic.ai: $0.25 per 1,000 requests Linkup: $5.00 per 1,000 requests Exa: $7.00 per 1,000 requests [0] https://serper.dev/ [1] https://developers.cloudflare.com/web-search/providers/", "context": "", "use_case": "", "switched_from": "", "story": "Web Search API"}, {"id": 792790528, "item_id": 49549408, "url": "https://news.ycombinator.com/item?id=49549408", "by": "mark_l_watson", "date": "2026-09-03", "version": null, "stance": "neutral", "quote": "I use Exa and Brave search APIs separately", "excerpt": "That is my understanding, which is a nice thing. I use Exa and Brave search APIs separately but the Google bundling is convenient. Ollama Cloud offers the same thing: they supply a web search tool bundled with cloud API inference services.", "context": "", "use_case": "search API for LLMs", "switched_from": "", "story": "Gemini 3.8 Flash and 3.8 Flash Cyber"}, {"id": 792424352, "item_id": 49526522, "url": "https://news.ycombinator.com/item?id=49526522", "by": "verdverm", "date": "2026-09-01", "version": null, "stance": "recommend", "quote": "try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets.", "excerpt": "My experience is otherwise, try alternative search SaaS like Exa, i.e. pay per search, but use multiple because they return different sets. My agent looks at the results, includes some, performs more searches. It's more of a deep search thing, very similar in spirit to how I used to run multiple queries, click a bunch of links, evaluate, keep good ones, run follow up queries, and end up with an alt-tab of dozens of b", "context": "", "use_case": "agent web search", "switched_from": "", "story": "Thanks to Lake Ontario, MapQuest is popular all over again"}, {"id": 791310176, "item_id": 49456886, "url": "https://news.ycombinator.com/item?id=49456886", "by": "usef-", "date": "2026-08-26", "version": null, "stance": "recommend", "quote": "Exa also has an API for it that has worked well for me", "excerpt": "Exa also has an API for it that has worked well for me, returning markdown for a URL, which means you don't need to render js or anything yourself. It doesn't need an account for up to 1k requests/month, which is more than I've ever needed. Most harnesses have a \"zero config\" plugin that works, eg: https://pi.dev/packages/pi-exa (no affiliation to either)", "context": "", "use_case": "Fetching URLs as markdown", "switched_from": "", "story": "Serve Markdown to AI Agents with Accept Headers"}, {"id": 790983073, "item_id": 49436442, "url": "https://news.ycombinator.com/item?id=49436442", "by": "phoenix120", "date": "2026-08-25", "version": null, "stance": "compare", "quote": "Exa charges 7$ per 1000 queries, and charges additionally for each result that is out of top-10.", "excerpt": "Some additional points to what Matthias already said in the same thread. The main differences are lower price and latency while having roughly the same (or sometime superior) quality. In terms of price Keenable is 4$ per 1000 queries, and for each query you can have up to 50 results for the same price. Exa charges 7$ per 1000 queries, and charges additionally for each result that is out of top-10. Parallel charges 5$", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Keenable \u2013 A different web search API for AI agents"}, {"id": 790982576, "item_id": 49436411, "url": "https://news.ycombinator.com/item?id=49436411", "by": "matt4711", "date": "2026-08-25", "version": null, "stance": "compare", "quote": "we are competitive in quality to exa (the market leader) but much cheaper and lower latency", "excerpt": "If you look at our benchmarks at https://keenableai.github.io/needle/ we are competitive in quality to exa (the market leader) but much cheaper and lower latency. This provides interesting tradeoffs where models can call search much more frequently which we believe is one thing that is not happening today. Our benchmarks (see https://keenableai.github.io/needle/card-lowshared/ ) also suggest that some players in the ", "context": "", "use_case": "", "switched_from": "", "story": "Show HN: Keenable \u2013 A different web search API for AI agents"}, {"id": 789621008, "item_id": 49351313, "url": "https://news.ycombinator.com/item?id=49351313", "by": "fwn", "date": "2026-08-18", "version": null, "stance": "recommend", "quote": "exa.ai/about looks interesting to me", "excerpt": "Mozilla PR department seems to be vanilla Claude. tl;dr of the article: Rather than building a closed AI stack (like Google - but they don't name it) Mozilla is pursuing partnerships. In this case, they are partnering with Exa.ai to offer AI integrations on desktop and iOS devices with visible source citations. There is no mention of confidential computing, hardware enclaves, or other technical mechanisms but they pr", "context": "", "use_case": "AI search API", "switched_from": "", "story": "Firefox and Exa Partnership"}, {"id": 789339041, "item_id": 49333690, "url": "https://news.ycombinator.com/item?id=49333690", "by": "yoeven", "date": "2026-08-17", "version": null, "stance": "compare", "quote": "Exa could be great in some for example people search but is pretty bad at getting the latest content of the page", "excerpt": "thanks! 1. yeah, we've optimized significantly for latency, give the API a try and you'll see it's pretty fast. 2. more than you think, we use a wide range of providers internally depending on the context needed. Exa could be great in some for example people search but is pretty bad at getting the latest content of the page, Parallel is better there. Apify is great for getting google ranked pages. Perplexity is great", "context": "", "use_case": "people search", "switched_from": "", "story": "Show HN: OpenWebSearch \u2013 A router for web search indexes"}, {"id": 787829088, "item_id": 49239318, "url": "https://news.ycombinator.com/item?id=49239318", "by": "ignoramous", "date": "2026-08-09", "version": null, "stance": "recommend", "quote": "see exa.ai who build generic solutions in similar space", "excerpt": "It is both a great (ai tools) & worst (ai slop) time to be building a web search engine! > working on some more commercial features, like the ability to monitor queries (e.g. for backlinks or brand mentions) and get an email when there are new results As an inspiration (presuming you don't already know), see exa.ai who build generic solutions in similar space: https://exa.ai/docs/reference/monitors-api-guide", "context": "", "use_case": "Search API", "switched_from": "", "story": "Ask HN: What are you working on? (August 2026)"}, {"id": 787698994, "item_id": 49231187, "url": "https://news.ycombinator.com/item?id=49231187", "by": "thepoet", "date": "2026-08-09", "version": null, "stance": "neutral", "quote": "generate audiobooks on topics I want to understand with ElevenLabs, Exa and OpenAI", "excerpt": "I have a Kobo that was kept unused in a similar state for a few years until I decide to build an SDK platform to develop and deploy apps on it. My Kobo does not support apps on its Linux platform. I have been using it to read HN, as an RSS reader over my news and blog sources, generate audiobooks on topics I want to understand with ElevenLabs, Exa and OpenAI and listen via connecting its bluetooth to my speakers and ", "context": "", "use_case": "search for audiobook content", "switched_from": "", "story": "Reviving a four year old reMarkable 2"}, {"id": 785212065, "item_id": 49075754, "url": "https://news.ycombinator.com/item?id=49075754", "by": "nijave", "date": "2026-07-27", "version": null, "stance": "recommend", "quote": "results got better when it modified SearXNG to work very similarly to Exa", "excerpt": "Forgot about the ol \"but how many tokens did you spend to get _there_\"--wish benchmarks would include the number of input/output tokens to achieve the score. I think the closest is Android Bench https://developer.android.com/bench although best you can do is extrapolate off time/cost (iirc they claim to prefer using provider's native API) In general, smart models work fine with any tools, dumber models need better to", "context": "", "use_case": "", "switched_from": "", "story": "Opus 5 is currently #1 on Artificial Analysis Intelligence Leaderboard"}, {"id": 781653969, "item_id": 48853373, "url": "https://news.ycombinator.com/item?id=48853373", "by": "nijave", "date": "2026-07-09", "version": null, "stance": "recommend", "quote": "Exa MCP tools generally worked better", "excerpt": "It's both. Some models, given web search and web fetch may only run a search and assume the summary text is correct and blindly return it. Others will validate by running a web fetch and checking the whole page contents. Even better, the model will run multiple searches and fetches to cross check the information. The best models will attempt to verify whether a source is authoritative or not and try to only return au", "context": "", "use_case": "Web search for LLMs", "switched_from": "", "story": "Grok 4.5"}, {"id": 781377584, "item_id": 48836099, "url": "https://news.ycombinator.com/item?id=48836099", "by": "nijave", "date": "2026-07-08", "version": null, "stance": "recommend", "quote": "Exa MCP annihilates them", "excerpt": "Arguably I'd call that the 90%. In my case, answering the restaurant question correctly with \"Rishi\" in my tests was the sole intent and 90% of the problem. All the models \"helpfully\" added extra junk about the closure, dates, quotes, etc and many of them got these details wrong--the 10% or extra crap not central to the question. If the central question was \"what is the bus schedule on `day`\" and the model screws tha", "context": "", "use_case": "Web search", "switched_from": "", "story": "Grok 4.5"}, {"id": 781066113, "item_id": 48816632, "url": "https://news.ycombinator.com/item?id=48816632", "by": "traceroute66", "date": "2026-07-07", "version": null, "stance": "recommend", "quote": "I have had success with exa[1]", "excerpt": "The blog author complains of \"lack of/poor web search capabilities\" in GLM, but you can always use it against an MCP of which there are many. For applications where I am not concerned about my queries being passed through a US provider, I have had success with exa[1] There are also other ways to give it context without web-search. For example the various MCPs that make `man` pages available. I've also found GLM to be", "context": "", "use_case": "Web search via MCP", "switched_from": "", "story": "GLM 5.2 and the coming AI margin collapse"}, {"id": 772261904, "item_id": 48266369, "url": "https://news.ycombinator.com/item?id=48266369", "by": "scratchyone", "date": "2026-05-25", "version": null, "stance": "recommend", "quote": "I've had generally good experiences with Exa", "excerpt": "I've had generally good experiences with Exa, I use it as an MCP in all my various coding tools and use the API whenever I build an AI thing that needs search. Also an extremely happy Kagi user but can't speak to their API, haven't tried it yet.", "context": "", "use_case": "Search API and MCP", "switched_from": "", "story": "Search engines alternatives now that Google isn't Google anymore"}, {"id": 772259552, "item_id": 48266222, "url": "https://news.ycombinator.com/item?id=48266222", "by": "smgpie", "date": "2026-05-25", "version": null, "stance": "compare", "quote": "I am thinking about either exa or kagi, but I have no idea which one is better.", "excerpt": "I am indeed looking into an alternative to Google recently, but more because of my need of a good search mcp server for my coding agents. I am thinking about either exa or kagi, but I have no idea which one is better. Also exa seems not quite frequently mentioned in the community, wondering why.", "context": "", "use_case": "search MCP server for coding agents", "switched_from": "", "story": "Search engines alternatives now that Google isn't Google anymore"}, {"id": 760239248, "item_id": 47514953, "url": "https://news.ycombinator.com/item?id=47514953", "by": "arthurjean", "date": "2026-03-25", "version": null, "stance": "warn", "quote": "Exa's MCP server is stateless, just a REST wrapper", "excerpt": "You're right actually. Exa's MCP server is stateless, just a REST wrapper. A skill + CLI would do the same job with way less context cost. Someone already built that (https://github.com/tobalsan/exa).", "context": "", "use_case": "", "switched_from": "", "story": "Welcome to FastMCP"}, {"id": 758716322, "item_id": 47419770, "url": "https://news.ycombinator.com/item?id=47419770", "by": "muggermuch", "date": "2026-03-17", "version": null, "stance": "neutral", "quote": "along with Exa via OpenRouter's integration", "excerpt": "> So someone monitoring Polymarket could have reached the same conclusion? Maybe? If they are professionally trading prediction markets, I'm pretty sure that would be the case. Polymarket especially is a great source of insider traded information, as you pointed out. We do near realtime tracking of most major markets, plus X accounts that Soros identifies as being important. The system also composes search queries pe", "context": "", "use_case": "search for analysis", "switched_from": "", "story": "Show HN: Soros \u2013 AI for geopolitical macro investing"}, {"id": 757407828, "item_id": 47337989, "url": "https://news.ycombinator.com/item?id=47337989", "by": "robthompson2018", "date": "2026-03-11", "version": null, "stance": "neutral", "quote": "I use Exa search to help me prepare for calls", "excerpt": "Our average user spends $50 a month all-in (tokens and subscription). If you're budget conscious you can use a cheap model (eg Gemini Flash) or even a free one. I confess I am a snob and only use Claude Opus, but even using OpenClaw all day every day I only spend about $500 a month on tokens. Orthogonal credits are used more frequently by power users. For everyday tasks they'll last a very long time, I don't think an", "context": "", "use_case": "Background research on prospects", "switched_from": "", "story": "Show HN: Klaus \u2013 OpenClaw on a VM, batteries included"}, {"id": 752056704, "item_id": 47003544, "url": "https://news.ycombinator.com/item?id=47003544", "by": "tibbar", "date": "2026-02-13", "version": null, "stance": "warn", "quote": "In practice, it's never worked that well for me", "excerpt": "Oh, like, I wouldn't actually use this specific implementation. I used to work at a shop with thousands of lines of Oracle triggers that you had to edit inline in the web browser with no version control and I shudder to think of returning again. I'm more interested in the data flow. exa.ai got famous for promising search with massively parallel execution of LLMs on candidate results. In practice, it's never worked th", "context": "", "use_case": "AI web search", "switched_from": "", "story": null}, {"id": 750242417, "item_id": 46890151, "url": "https://news.ycombinator.com/item?id=46890151", "by": "mheavers", "date": "2026-02-04", "version": null, "stance": "warn", "quote": "We switched from Exa because Tabstack proved more reliable at retrieving content from social media urls", "excerpt": "My team's product uses Tabstack's other structured data endpoints to retrieve the content of URLs for filling out of profile form fields and it has been flawless. We switched from Exa because Tabstack proved more reliable at retrieving content from social media urls", "context": "team building a product", "use_case": "retrieving URL content", "switched_from": "", "story": "Show HN: Tabstack Research \u2013 An API for verified web research (by Mozilla)"}, {"id": 747981776, "item_id": 46748861, "url": "https://news.ycombinator.com/item?id=46748861", "by": "-__-", "date": "2026-01-24", "version": null, "stance": "recommend", "quote": "exa.ai/search and perplexity", "excerpt": "exa.ai/search and perplexity, and most other search engines if you put the special characters in quotes. you could also find/make an instance of searx or search across aggregators. https://searx.space/", "context": "", "use_case": "", "switched_from": "", "story": "Ask HN: General purpose search engine that will respect special characters?"}, {"id": 747593665, "item_id": 46724604, "url": "https://news.ycombinator.com/item?id=46724604", "by": "leo_researchly", "date": "2026-01-22", "version": null, "stance": "recommend", "quote": "Firecrawl or Exa are pretty good for that", "excerpt": "We have been working on how to track Y Combinator Companies (and other incubators for that matter) and I wanted to share a few learnings: Identifying them in the first place is easy; even with a simple scraper you can save them to Notion, Google Sheets etc. Firecrawl or Exa are pretty good for that The hard part is keeping the list current and \"actionable\": you need a way to cluster startups; most VCs (& their thesis", "context": "", "use_case": "Web search and scraping", "switched_from": "", "story": "Why it's still hard to track Y Combinator Companies"}, {"id": 741660449, "item_id": 46353778, "url": "https://news.ycombinator.com/item?id=46353778", "by": "glenpk", "date": "2025-12-22", "version": null, "stance": "neutral", "quote": "Uses Exa (Semantic Search) to find libraries", "excerpt": "Hey everyone, The Problem: We\u2019ve all been there. You ask an LLM to \"solve a niche optimization problem.\" It writes the code beautifully, but it imports a library that hasn't been updated since 2018 or hallucinates an API entirely. You waste 30 minutes debugging only to realize you\u2019re trying to breathe life into a zombie repository. The Solution: I\u2019m a lawyer and psychologist by training\u2014essentially a professional \"wo", "context": "", "use_case": "Semantic search", "switched_from": "", "story": "Show HN: Context Agent\" for LLMs coding libraries (lawyer turned dev project)"}]}