My initial motivation is that I want a graph database. Lightweight and easy to use, like SQLite. Powerful and performant, like Postgres. I found none of the existing solutions good enough.
Deciding to roll my own, I need to choose a query language. I am familiar with Cypher but consider it not much of an improvement over CTE in SQL (Cypher is sometimes notationally more convenient, but not more expressive). I like Gremlin but would prefer something more declarative. Experimentations with Datomic and its clones convinced me that Datalog is the way to go.
Then I need a data model. I find the property graph model (Neo4j, etc.) over-constraining, and the triple store model (Datomic, etc.) suffering from inherent performance problems. They also lack the most important property of the relational model: being an algebra. Non-algebraic models are not very composable: you may store data as property graphs or triples, but when you do a query, you always get back relations. So I decided to have relational algebra as the data model.
The end result, I now present to you. Let me know what you think, good or bad, and I'll do my best to address them. This is the first time that I use Rust in a significant project, and I love the experience!"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Show HN: Cozo \u2013 new Graph DB with Datalog, embedded like SQLite"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/cozodb/cozo"}},"_tags":["story","author_zh217","story_33518320","show_hn"],"author":"zh217","children":[33518701,33518744,33518792,33518820,33519118,33519341,33519482,33520377,33520583,33520718,33521212,33521543,33521638,33522353,33522521,33522804,33523805,33525040,33526673,33528940,33529607],"created_at":"2022-11-08T12:25:23Z","created_at_i":1667910323,"num_comments":67,"objectID":"33518320","points":425,"story_id":33518320,"story_text":"Hi HN, I have been making this Cozo database since half a year ago, and now it is ready for public release.
My initial motivation is that I want a graph database. Lightweight and easy to use, like SQLite. Powerful and performant, like Postgres. I found none of the existing solutions good enough.
Deciding to roll my own, I need to choose a query language. I am familiar with Cypher but consider it not much of an improvement over CTE in SQL (Cypher is sometimes notationally more convenient, but not more expressive). I like Gremlin but would prefer something more declarative. Experimentations with Datomic and its clones convinced me that Datalog is the way to go.
Then I need a data model. I find the property graph model (Neo4j, etc.) over-constraining, and the triple store model (Datomic, etc.) suffering from inherent performance problems. They also lack the most important property of the relational model: being an algebra. Non-algebraic models are not very composable: you may store data as property graphs or triples, but when you do a query, you always get back relations. So I decided to have relational algebra as the data model.
The end result, I now present to you. Let me know what you think, good or bad, and I'll do my best to address them. This is the first time that I use Rust in a significant project, and I love the experience!","title":"Show HN: Cozo \u2013 new Graph DB with Datalog, embedded like SQLite","updated_at":"2026-05-06T15:13:12Z","url":"https://github.com/cozodb/cozo"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"epeus"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Twitter is using its embedded JavaScript to hide tweets that have been deleted"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://www.kevinmarks.com/twittereditsyou.html"}},"_tags":["story","author_epeus","story_30928217"],"author":"epeus","children":[30928264,30928268,30929023,30929073,30929076,30929164,30929177,30929191,30929236,30929271,30929280,30929335,30929355,30929378,30929421,30929432,30929462,30929524,30929546,30929559,30929613,30929636,30929674,30929718,30929719,30929724,30929793,30929817,30930255,30930397,30930490,30930497,30930536,30930738,30931080,30931312,30931365,30931873,30932912,30935169,30935300,30935755,30945579],"created_at":"2022-04-06T03:38:00Z","created_at_i":1649216280,"num_comments":245,"objectID":"30928217","points":408,"story_id":30928217,"title":"Twitter is using its embedded JavaScript to hide tweets that have been deleted","updated_at":"2024-09-20T10:51:07Z","url":"http://www.kevinmarks.com/twittereditsyou.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"formalsystem"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"The Embedded Rust Book"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://rust-embedded.github.io/book/intro/index.html"}},"_tags":["story","author_formalsystem","story_19766713"],"author":"formalsystem","children":[19767112,19767132,19767557,19767652,19767803,19767970,19768111,19768118,19768122,19768130,19768161,19768437],"created_at":"2019-04-27T17:49:08Z","created_at_i":1556387348,"num_comments":82,"objectID":"19766713","points":387,"story_id":19766713,"title":"The Embedded Rust Book","updated_at":"2024-09-20T04:07:16Z","url":"https://rust-embedded.github.io/book/intro/index.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"jyooru"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Git-bug: Distributed, offline-first bug tracker embedded in Git"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/MichaelMure/git-bug"}},"_tags":["story","author_jyooru","story_33730417"],"author":"jyooru","children":[33730900,33730922,33730969,33730972,33731118,33731681,33731870,33732047,33732064,33732262,33732438,33732470,33732493,33732985,33733539,33733588,33734698,33734740,33735143,33735499,33735556],"created_at":"2022-11-24T12:11:48Z","created_at_i":1669291908,"num_comments":79,"objectID":"33730417","points":380,"story_id":33730417,"title":"Git-bug: Distributed, offline-first bug tracker embedded in Git","updated_at":"2025-07-04T22:02:32Z","url":"https://github.com/MichaelMure/git-bug"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"timsneath"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Graphical User Interface Using Flutter in Embedded Systems [pdf]"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://static.sched.com/hosted_files/osseu2020/e0/Oct%2027_Graphical%20User%20Interface%20Using%20Flutter%20in%20Embedded%20Systems_Hidenori%20Matsubayashi.pdf"}},"_tags":["story","author_timsneath","story_24921031"],"author":"timsneath","children":[24921801,24921820,24921922,24921978,24922013,24922024,24922180,24922191,24922206,24922234,24922344,24922549,24922559,24922751,24922780,24922819,24923234,24923285,24923335,24923422,24923462,24923480,24923569,24923784,24923920,24924142,24924212,24924294,24924334,24924431,24924572,24925214,24925313,24925483,24928548],"created_at":"2020-10-28T16:49:56Z","created_at_i":1603903796,"num_comments":197,"objectID":"24921031","points":378,"story_id":24921031,"title":"Graphical User Interface Using Flutter in Embedded Systems [pdf]","updated_at":"2025-04-10T19:38:35Z","url":"https://static.sched.com/hosted_files/osseu2020/e0/Oct%2027_Graphical%20User%20Interface%20Using%20Flutter%20in%20Embedded%20Systems_Hidenori%20Matsubayashi.pdf"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"1900jwatson"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"AMD Launches Ultra-Low-Power Ryzen Embedded APUs"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://www.anandtech.com/show/15554/amd-launches-ultralowpower-ryzen-embedded-apus-starting-at-6w"}},"_tags":["story","author_1900jwatson","story_22440894"],"author":"1900jwatson","children":[22441004,22441010,22441023,22441078,22441144,22442031,22442091,22442174,22442569,22442694,22443081,22443280,22443494,22444726,22445093,22450069],"created_at":"2020-02-28T06:19:00Z","created_at_i":1582870740,"num_comments":144,"objectID":"22440894","points":373,"story_id":22440894,"title":"AMD Launches Ultra-Low-Power Ryzen Embedded APUs","updated_at":"2024-12-01T21:08:06Z","url":"https://www.anandtech.com/show/15554/amd-launches-ultralowpower-ryzen-embedded-apus-starting-at-6w"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"deepsquirrelnet"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"After working with LLMs for long enough, I found myself wanting a lightweight utility for doing various small tasks to prepare inputs, locate information and create evaluators. This library is two things: a very simple model and utilities that inference it (eg. fuzzy deduplication). The target platform is CPU, and it\u2019s intended to be light, fast and pip installable \u2014 a library that lowers the barrier to working with strings semantically. You don\u2019t need to install pytorch to use it, or any deep learning runtimes.
How can this be accomplished? The model is simply token embeddings that are average pooled. To create this model, I extracted token embedding (nn.Embedding) vectors from LLMs, concatenated them along the embedding dimension, added a learnable weight parameter, and projected them to a smaller dimension. Using the sentence transformers framework and datasets, I trained the pooled embedding with multiple negatives ranking loss and matryoshka representation learning so they can be truncated. After training, the weights and projections are no longer needed, because there is no contextual calculations. I inference the entire token vocabulary and save the new token embeddings to be loaded to numpy.
While the results are not impressive compared to transformer models, they perform well on MTEB benchmarks compared to word embedding models (which they are most similar to), while being much smaller in size (smallest model, 32k vocab, 64-dim is only 4MB).
On the utility side, I\u2019ve been adding some tools that I think it\u2019ll be useful for. In addition to general embedding, there\u2019s algorithms for ranking, filtering, clustering, deduplicating and similarity. Some of them have a cython implementation, and I\u2019m continuing to work on benchmarking them and improving them as I have time. In addition to \u201cstandard\u201d models that use cosine similarity for some algorithms, there are binarized models that use hamming distance. This is a slightly faster, similarity algorithm, with significantly less memory per embedding (float32 -> 1 bit).
Hope you enjoy it, and find it useful. PS I haven\u2019t figured out Windows builds yet, but Linux and Mac are supported."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Show HN: Wordllama \u2013 Things you can do with the token embeddings of an LLM"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/dleemiller/WordLlama"}},"_tags":["story","author_deepsquirrelnet","story_41544969","show_hn"],"author":"deepsquirrelnet","children":[41545666,41545765,41545767,41546377,41547225,41547680,41548377,41548740,41552424,41553869],"created_at":"2024-09-15T03:25:14Z","created_at_i":1726370714,"num_comments":36,"objectID":"41544969","points":370,"story_id":41544969,"story_text":"After working with LLMs for long enough, I found myself wanting a lightweight utility for doing various small tasks to prepare inputs, locate information and create evaluators. This library is two things: a very simple model and utilities that inference it (eg. fuzzy deduplication). The target platform is CPU, and it\u2019s intended to be light, fast and pip installable \u2014 a library that lowers the barrier to working with strings semantically. You don\u2019t need to install pytorch to use it, or any deep learning runtimes.
How can this be accomplished? The model is simply token embeddings that are average pooled. To create this model, I extracted token embedding (nn.Embedding) vectors from LLMs, concatenated them along the embedding dimension, added a learnable weight parameter, and projected them to a smaller dimension. Using the sentence transformers framework and datasets, I trained the pooled embedding with multiple negatives ranking loss and matryoshka representation learning so they can be truncated. After training, the weights and projections are no longer needed, because there is no contextual calculations. I inference the entire token vocabulary and save the new token embeddings to be loaded to numpy.
While the results are not impressive compared to transformer models, they perform well on MTEB benchmarks compared to word embedding models (which they are most similar to), while being much smaller in size (smallest model, 32k vocab, 64-dim is only 4MB).
On the utility side, I\u2019ve been adding some tools that I think it\u2019ll be useful for. In addition to general embedding, there\u2019s algorithms for ranking, filtering, clustering, deduplicating and similarity. Some of them have a cython implementation, and I\u2019m continuing to work on benchmarking them and improving them as I have time. In addition to \u201cstandard\u201d models that use cosine similarity for some algorithms, there are binarized models that use hamming distance. This is a slightly faster, similarity algorithm, with significantly less memory per embedding (float32 -> 1 bit).
Hope you enjoy it, and find it useful. PS I haven\u2019t figured out Windows builds yet, but Linux and Mac are supported.","title":"Show HN: Wordllama \u2013 Things you can do with the token embeddings of an LLM","updated_at":"2026-08-05T14:54:19Z","url":"https://github.com/dleemiller/WordLlama"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"misonic"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embeddings are underrated"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://technicalwriting.dev/data/embeddings.html"}},"_tags":["story","author_misonic","story_42013762"],"author":"misonic","children":[42014036,42014125,42014173,42014495,42014634,42014683,42014831,42014850,42014936,42015069,42015282,42015484,42015723,42015790,42015899,42016369,42016494,42016650,42016671,42016747,42016799,42018485,42018957,42019068,42019380,42020122,42020927,42022496,42025726,42027944,42029244,42030692],"created_at":"2024-11-01T03:24:17Z","created_at_i":1730431457,"num_comments":176,"objectID":"42013762","points":367,"story_id":42013762,"title":"Embeddings are underrated","updated_at":"2025-11-02T05:12:34Z","url":"https://technicalwriting.dev/data/embeddings.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"kartD"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embedded in Rust: Brave new I/O"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://blog.japaric.io/brave-new-io/"}},"_tags":["story","author_kartD","story_16189088"],"author":"kartD","children":[16189570,16189710,16191011,16191198,16191885,16192547,16193142,16193331,16199292],"created_at":"2018-01-19T19:41:59Z","created_at_i":1516390919,"num_comments":65,"objectID":"16189088","points":365,"story_id":16189088,"title":"Embedded in Rust: Brave new I/O","updated_at":"2024-09-20T01:57:34Z","url":"http://blog.japaric.io/brave-new-io/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"danso"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Tracking leaks on NDA beta Xbox 360s by embedding serial number on-screen"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://twitter.com/cullend/status/1071884772064944128"}},"_tags":["story","author_danso","story_18643987"],"author":"danso","children":[18648903,18648973,18649015,18649033,18649054,18649066,18649071,18649080,18649338,18649366,18649701,18649898,18650000,18650508,18650576,18650630,18650753,18651269,18652244,18654057],"created_at":"2018-12-09T23:25:44Z","created_at_i":1544397944,"num_comments":140,"objectID":"18643987","points":357,"story_id":18643987,"title":"Tracking leaks on NDA beta Xbox 360s by embedding serial number on-screen","updated_at":"2024-09-20T03:30:11Z","url":"https://twitter.com/cullend/status/1071884772064944128"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dredmorbius"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Starting January 4 Google will block sign-ins from embedded browser frameworks"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://lists.webkit.org/pipermail/webkit-dev/2020-November/031604.html"}},"_tags":["story","author_dredmorbius","story_25155451"],"author":"dredmorbius","children":[25155538,25155568,25155569,25155583,25155608,25155612,25155623,25155628,25155631,25155646,25155665,25155669,25155685,25155712,25155713,25155716,25155718,25155720,25155730,25155785,25155870,25155954,25155987,25156071,25156137,25156722,25157038,25157128,25157301,25157402,25174703,25175566],"created_at":"2020-11-19T23:07:58Z","created_at_i":1605827278,"num_comments":186,"objectID":"25155451","points":326,"story_id":25155451,"title":"Starting January 4 Google will block sign-ins from embedded browser frameworks","updated_at":"2024-09-20T07:19:07Z","url":"https://lists.webkit.org/pipermail/webkit-dev/2020-November/031604.html"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"soycaporal"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Ternlight \u2013 7 MB embedding model that runs in browser (WASM)"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://ternlight-demo.vercel.app/"}},"_tags":["story","author_soycaporal","story_48811644"],"author":"soycaporal","children":[48811645,48812044,48812070,48812124,48812176,48812240,48812306,48812340,48812345,48812584,48812772,48813151,48813285,48813533,48813838,48813964,48814357,48814371,48814662,48815373,48815977,48815988,48816530,48816746,48816750,48817257,48817414,48817858,48832934,48836736],"created_at":"2026-07-06T23:06:31Z","created_at_i":1783379191,"num_comments":70,"objectID":"48811644","points":326,"story_id":48811644,"title":"Ternlight \u2013 7 MB embedding model that runs in browser (WASM)","updated_at":"2026-09-06T08:35:24Z","url":"https://ternlight-demo.vercel.app/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"fniephaus"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"GraalPy \u2013 A high-performance embeddable Python 3 runtime for Java"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://www.graalvm.org/python/"}},"_tags":["story","author_fniephaus","story_41570708"],"author":"fniephaus","children":[41571052,41571064,41571220,41571544,41571576,41571632,41571678,41571707,41571710,41571742,41571786,41571900,41572436,41573054,41573105,41573801,41573814,41573905,41574419,41575889,41576536,41576608,41578690,41579186],"created_at":"2024-09-17T18:04:31Z","created_at_i":1726596271,"num_comments":141,"objectID":"41570708","points":325,"story_id":41570708,"title":"GraalPy \u2013 A high-performance embeddable Python 3 runtime for Java","updated_at":"2026-05-09T20:35:38Z","url":"https://www.graalvm.org/python/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tonyskn"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"I solved the embedded Chrome OS ad equation and won a Cr-48"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://www.sylvainzimmer.com/2010/12/10/first-to-decode-the-chrome-os-video-equation-won-a-cr-48/"}},"_tags":["story","author_tonyskn","story_1992804"],"author":"tonyskn","children":[1992943,1992944,1992965,1993007,1993014,1993031,1993066,1993265,1993300,1993342,1993646,1993655,1993908,1994060,2017441],"created_at":"2010-12-10T19:28:47Z","created_at_i":1292009327,"num_comments":88,"objectID":"1992804","points":324,"story_id":1992804,"title":"I solved the embedded Chrome OS ad equation and won a Cr-48","updated_at":"2026-05-11T00:08:41Z","url":"http://www.sylvainzimmer.com/2010/12/10/first-to-decode-the-chrome-os-video-equation-won-a-cr-48/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"birdculture"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embassy: Modern embedded framework, using Rust and async"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/embassy-rs/embassy"}},"_tags":["story","author_birdculture","story_46547740"],"author":"birdculture","children":[46547894,46547970,46548097,46548122,46548168,46548458,46548899,46548963,46549059,46549138,46549144,46549272,46549630,46550692,46551168,46551369,46552464,46553141,46553143,46553226,46553236,46553788,46554294,46554640,46556318],"created_at":"2026-01-08T23:00:45Z","created_at_i":1767913245,"num_comments":161,"objectID":"46547740","points":323,"story_id":46547740,"title":"Embassy: Modern embedded framework, using Rust and async","updated_at":"2026-08-01T03:12:35Z","url":"https://github.com/embassy-rs/embassy"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"colinprince"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Hy \u2013\u00a0A dialect of Lisp that\u2019s embedded in Python"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://docs.hylang.org/en/latest/"}},"_tags":["story","author_colinprince","story_8696975"],"author":"colinprince","children":[8697074,8697095,8697105,8697115,8697133,8697163,8697191,8697200,8697321,8697357,8697391,8697551,8697618,8698282,8698665,8698984,8699032,8699069,8699510,8699521,8699802,8699914],"created_at":"2014-12-03T23:14:55Z","created_at_i":1417648495,"num_comments":77,"objectID":"8696975","points":313,"story_id":8696975,"title":"Hy \u2013\u00a0A dialect of Lisp that\u2019s embedded in Python","updated_at":"2024-11-12T00:40:37Z","url":"http://docs.hylang.org/en/latest/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"stefankuehnel"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Git Bug: Distributed, Offline-First Bug Tracker Embedded in Git, with Bridges"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/git-bug/git-bug"}},"_tags":["story","author_stefankuehnel","story_43971620"],"author":"stefankuehnel","children":[43986409,43986717,43986822,43987043,43987623,43987656,43987744,43987754,43987879,43988552,43989310,43989810,43989811,43990074,43990123,43990347,43990787,43992360,43992674,43992725,43992761,43992796,43994285,43994947,43996550,43996693,43997314],"created_at":"2025-05-13T11:07:29Z","created_at_i":1747134449,"num_comments":107,"objectID":"43971620","points":312,"story_id":43971620,"title":"Git Bug: Distributed, Offline-First Bug Tracker Embedded in Git, with Bridges","updated_at":"2026-08-30T15:50:34Z","url":"https://github.com/git-bug/git-bug"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"GavinAnderegg"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Blocking Visual Studio Code embedded reverse shell before it's too late"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://ipfyx.fr/post/visual-studio-code-tunnel/"}},"_tags":["story","author_GavinAnderegg","story_37623562"],"author":"GavinAnderegg","children":[37623979,37624037,37624063,37624115,37624132,37624159,37624329,37624429,37624543,37624594,37625320,37625402,37625471,37625568,37625796,37630269],"created_at":"2023-09-23T14:29:55Z","created_at_i":1695479395,"num_comments":83,"objectID":"37623562","points":310,"story_id":37623562,"title":"Blocking Visual Studio Code embedded reverse shell before it's too late","updated_at":"2026-04-25T08:38:04Z","url":"https://ipfyx.fr/post/visual-studio-code-tunnel/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"arathunku"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embedding Python in Elixir, it's fine"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://dashbit.co/blog/running-python-in-elixir-its-fine"}},"_tags":["story","author_arathunku","story_43171239"],"author":"arathunku","children":[43171907,43172090,43172211,43172284,43172325,43172445,43173070,43173149,43173176,43173204,43173264,43173491,43173546,43173698,43174359,43177285,43179534],"created_at":"2025-02-25T12:53:15Z","created_at_i":1740487995,"num_comments":64,"objectID":"43171239","points":308,"story_id":43171239,"title":"Embedding Python in Elixir, it's fine","updated_at":"2025-07-09T10:22:04Z","url":"https://dashbit.co/blog/running-python-in-elixir-its-fine"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"goranmoomin"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embedded Swift on the Raspberry Pi Pico"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://forums.swift.org/t/embedded-swift-on-the-raspberry-pi-pico-rp2040-without-the-pico-sdk/69338"}},"_tags":["story","author_goranmoomin","story_39611111"],"author":"goranmoomin","children":[39611261,39611426,39611559,39613244,39614234,39614372,39615988,39620132],"created_at":"2024-03-06T01:10:49Z","created_at_i":1709687449,"num_comments":153,"objectID":"39611111","points":293,"story_id":39611111,"title":"Embedded Swift on the Raspberry Pi Pico","updated_at":"2026-05-23T17:20:44Z","url":"https://forums.swift.org/t/embedded-swift-on-the-raspberry-pi-pico-rp2040-without-the-pico-sdk/69338"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ogoffart"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Slint is a declarative GUI toolkit written primarily in Rust, with API support for multiple programming languages such as C++ and JavaScript. It is designed for desktop and embedded usage.
The Slint website has just been redesigned. And we added a new Royalty-Free License option besides GPL and commercial"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Show HN: Slint \u2013 A declarative UI toolkit for embedded and desktop"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://slint.dev"}},"_tags":["story","author_ogoffart","story_36388821","show_hn"],"author":"ogoffart","children":[36388980,36389242,36389259,36389328,36389393,36389433,36389489,36389552,36389619,36389748,36389839,36389900,36390192,36390247,36390693,36390733,36391212,36391319,36391363,36391578,36392213,36393172,36393318,36393675,36393851,36395105,36395308,36395873,36396110,36396241,36396499,36397128,36397890,36399064],"created_at":"2023-06-19T08:57:22Z","created_at_i":1687165042,"num_comments":148,"objectID":"36388821","points":291,"story_id":36388821,"story_text":"Slint is a declarative GUI toolkit written primarily in Rust, with API support for multiple programming languages such as C++ and JavaScript. It is designed for desktop and embedded usage.
The Slint website has just been redesigned. And we added a new Royalty-Free License option besides GPL and commercial","title":"Show HN: Slint \u2013 A declarative UI toolkit for embedded and desktop","updated_at":"2026-03-15T00:11:00Z","url":"https://slint.dev"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tomduncalf"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Patching an embedded synthesiser OS from 1996 with Ghidra"},"url":{"matchLevel":"none","matchedWords":[],"value":"http://blog.petersobot.com/patching-the-k2500"}},"_tags":["story","author_tomduncalf","story_31144106"],"author":"tomduncalf","children":[31144450,31144713,31144955,31145132,31145142,31145367,31146537,31146775,31147354,31147756,31147788,31148940,31149105,31161215],"created_at":"2022-04-24T14:05:17Z","created_at_i":1650809117,"num_comments":32,"objectID":"31144106","points":289,"story_id":31144106,"title":"Patching an embedded synthesiser OS from 1996 with Ghidra","updated_at":"2024-09-20T10:58:12Z","url":"http://blog.petersobot.com/patching-the-k2500"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"simonpure"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Gemini Embedding: Powering RAG and context engineering"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://developers.googleblog.com/en/gemini-embedding-powering-rag-context-engineering/"}},"_tags":["story","author_simonpure","story_44747457"],"author":"simonpure","children":[44747991,44748089,44748266,44748272,44748335,44748695,44749167,44749352,44750258,44750523,44750835,44751712,44752788,44753925,44754078,44754986,44754993,44755462,44755586,44758417],"created_at":"2025-07-31T16:47:54Z","created_at_i":1753980474,"num_comments":92,"objectID":"44747457","points":278,"story_id":44747457,"title":"Gemini Embedding: Powering RAG and context engineering","updated_at":"2026-01-13T09:03:18Z","url":"https://developers.googleblog.com/en/gemini-embedding-powering-rag-context-engineering/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"mountainview"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"An embedded database written in Rust"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/spacejam/sled"}},"_tags":["story","author_mountainview","story_17170733"],"author":"mountainview","children":[17171016,17171043,17171074,17171095,17171194,17171363,17171454,17171642,17172840],"created_at":"2018-05-28T05:32:28Z","created_at_i":1527485548,"num_comments":72,"objectID":"17170733","points":277,"story_id":17170733,"title":"An embedded database written in Rust","updated_at":"2026-05-12T21:24:32Z","url":"https://github.com/spacejam/sled"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"maxloh"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Boa: A standard-conforming embeddable JavaScript engine written in Rust"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/boa-dev/boa"}},"_tags":["story","author_maxloh","story_45832051"],"author":"maxloh","children":[45939208,45939345,45939367,45939503,45939611,45939744,45939947,45939970,45939993,45942158,45942704,45944773,45948206],"created_at":"2025-11-06T06:15:43Z","created_at_i":1762409743,"num_comments":68,"objectID":"45832051","points":272,"story_id":45832051,"title":"Boa: A standard-conforming embeddable JavaScript engine written in Rust","updated_at":"2026-09-08T20:17:48Z","url":"https://github.com/boa-dev/boa"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"pykello"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"OpenWrt: A Linux OS targeting embedded devices"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://openwrt.org/"}},"_tags":["story","author_pykello","story_45170087"],"author":"pykello","children":[45170518,45170671,45170912,45171194,45171662,45171756,45172064,45172139,45172165,45172389,45172482,45172627,45172755,45172768,45172817,45172851,45172923,45173149,45173485,45174062,45174272,45174601,45175104,45175824,45175929,45176032,45177147,45196121],"created_at":"2025-09-08T16:13:10Z","created_at_i":1757347990,"num_comments":185,"objectID":"45170087","points":271,"story_id":45170087,"title":"OpenWrt: A Linux OS targeting embedded devices","updated_at":"2026-04-29T22:45:53Z","url":"https://openwrt.org/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"bealuga"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Hi Hacker News!
My name is Bea, I built a site called Libraria that uses GPT to do a few things
1. Let you spin up multiple assistants based on your own documents. You can make it public, private, or protected. It has its own subdomain and landing page.\n2. Respond in full markdown always, so it can output images, links, code, and more\n3. Let you upload articles on the fly within the Chat, so you can ask it questions\n4. Make it embeddable in your site with one line of code\n5. Let you update it for fun / with your branding\n5. Enable syncing for any URLs you let us scrape, so that you can make sure it's always up to date\n6. Let you upload multiple file types
I've been working on this for about a month now by myself and you can keep track of my feature updates here: https://libraria.dev/feature-updates
I would LOVE your feedback on anything, and If you're willing to try it out I'm looking for a few beta users that can provide me more continuous feedback that I would gladly waive the fee for!"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Show HN: Customizable, embeddable Chat GPT based on your own documents"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://libraria.dev/"}},"_tags":["story","author_bealuga","story_35350662","show_hn"],"author":"bealuga","children":[35350663,35350774,35351098,35351105,35351262,35351374,35351465,35351490,35351505,35351514,35351554,35351563,35351640,35351646,35351678,35351682,35351767,35351806,35351830,35351904,35352091,35352106,35352122,35352155,35352193,35352271,35352308,35352558,35352575,35352586,35352654,35352928,35353018,35353048,35353089,35353146,35353323,35353512,35353528,35353559,35353581,35353739,35353838,35353842,35353854,35354453,35355325,35355640,35355907,35356756,35356806,35356811,35357030,35357306,35357432,35357437,35358622,35361529,35375310,35377920],"created_at":"2023-03-29T01:11:18Z","created_at_i":1680052278,"num_comments":130,"objectID":"35350662","points":271,"story_id":35350662,"story_text":"Hi Hacker News!
My name is Bea, I built a site called Libraria that uses GPT to do a few things
1. Let you spin up multiple assistants based on your own documents. You can make it public, private, or protected. It has its own subdomain and landing page.\n2. Respond in full markdown always, so it can output images, links, code, and more\n3. Let you upload articles on the fly within the Chat, so you can ask it questions\n4. Make it embeddable in your site with one line of code\n5. Let you update it for fun / with your branding\n5. Enable syncing for any URLs you let us scrape, so that you can make sure it's always up to date\n6. Let you upload multiple file types
I've been working on this for about a month now by myself and you can keep track of my feature updates here: https://libraria.dev/feature-updates
I would LOVE your feedback on anything, and If you're willing to try it out I'm looking for a few beta users that can provide me more continuous feedback that I would gladly waive the fee for!","title":"Show HN: Customizable, embeddable Chat GPT based on your own documents","updated_at":"2024-09-20T13:38:54Z","url":"https://libraria.dev/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"punnerud"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Nerves \u2013 Craft and deploy bulletproof embedded software in Elixir"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://nerves-project.org"}},"_tags":["story","author_punnerud","story_16915337"],"author":"punnerud","children":[16915631,16915786,16916286,16916585,16917635,16917836,16920563,16922162],"created_at":"2018-04-24T18:36:12Z","created_at_i":1524594972,"num_comments":74,"objectID":"16915337","points":270,"story_id":16915337,"title":"Nerves \u2013 Craft and deploy bulletproof embedded software in Elixir","updated_at":"2024-09-20T02:22:22Z","url":"https://nerves-project.org"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"dannyobrien"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Berry is a ultra-lightweight dynamically typed embedded scripting language"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://berry-lang.github.io/"}},"_tags":["story","author_dannyobrien","story_37801140"],"author":"dannyobrien","children":[37801707,37801791,37801933,37802096,37802711,37802930,37803102,37803516,37803952,37804182,37804331,37804338,37804573,37804575,37804815,37804955,37805004,37805313,37805380,37805544,37805843,37809185,37809920,37810467],"created_at":"2023-10-07T12:26:57Z","created_at_i":1696681617,"num_comments":97,"objectID":"37801140","points":266,"story_id":37801140,"title":"Berry is a ultra-lightweight dynamically typed embedded scripting language","updated_at":"2025-10-30T13:54:26Z","url":"https://berry-lang.github.io/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"fzliu"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"All-in-one embedding model for interleaved text, images, and screenshots"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://blog.voyageai.com/2024/11/12/voyage-multimodal-3/"}},"_tags":["story","author_fzliu","story_42162622"],"author":"fzliu","children":[42163036,42163393,42163723,42163898,42163961,42164014,42164036,42164083,42164182,42166275,42167543,42181245,42193602],"created_at":"2024-11-17T07:42:08Z","created_at_i":1731829328,"num_comments":31,"objectID":"42162622","points":263,"story_id":42162622,"title":"All-in-one embedding model for interleaved text, images, and screenshots","updated_at":"2025-08-14T18:15:28Z","url":"https://blog.voyageai.com/2024/11/12/voyage-multimodal-3/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"hackerpain"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"The embedded YouTube player told me what you were watching"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://bugs.xdavidhu.me/google/2021/01/18/the-embedded-youtube-player-told-me-what-you-were-watching-and-more/"}},"_tags":["story","author_hackerpain","story_25823048"],"author":"hackerpain","children":[25823882,25824007,25824019,25824432,25824729,25825303,25825335,25826734,25827168,25827256,25827850,25828100,25828202],"created_at":"2021-01-18T16:02:34Z","created_at_i":1610985754,"num_comments":104,"objectID":"25823048","points":260,"story_id":25823048,"title":"The embedded YouTube player told me what you were watching","updated_at":"2024-09-20T07:45:52Z","url":"https://bugs.xdavidhu.me/google/2021/01/18/the-embedded-youtube-player-told-me-what-you-were-watching-and-more/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"ingve"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"Embedding Binary Objects in C"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://flak.tedunangst.com/post/embedding-binary-objects-in-c"}},"_tags":["story","author_ingve","story_22888318"],"author":"ingve","children":[22888425,22888445,22888532,22888568,22888677,22888843,22888854,22888872,22888960,22889048,22889066,22889143,22889275,22889326,22889346,22889504,22890020,22890268,22890306,22890434,22890507,22890784,22891002,22891281,22891330,22891911,22892526,22893634,22897071],"created_at":"2020-04-16T11:27:35Z","created_at_i":1587036455,"num_comments":136,"objectID":"22888318","points":257,"story_id":22888318,"title":"Embedding Binary Objects in C","updated_at":"2025-10-10T09:27:45Z","url":"https://flak.tedunangst.com/post/embedding-binary-objects-in-c"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Anon84"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"What are embeddings?"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["embedd"],"value":"https://vickiboykis.com/what_are_embeddings/"}},"_tags":["story","author_Anon84","story_36469910"],"author":"Anon84","children":[36470326,36470351,36470578,36470719,36470818,36472607,36473697,36475492,36476576,36478132,36479169],"created_at":"2023-06-25T16:27:06Z","created_at_i":1687710426,"num_comments":51,"objectID":"36469910","points":257,"story_id":36469910,"title":"What are embeddings?","updated_at":"2025-09-21T20:52:36Z","url":"https://vickiboykis.com/what_are_embeddings/"}],"hitsPerPage":50,"nbHits":14687,"nbPages":20,"page":0,"params":"query=Embedd&tags=story&hitsPerPage=50&advancedSyntax=true&analyticsTags=backend","processingTimeMS":13,"processingTimingsMS":{"_request":{"queue":7,"roundTrip":21},"afterFetch":{"format":{"total":1}},"fetch":{"query":3,"scanning":8,"total":12},"total":13},"query":"Embedd","serverTimeMS":21}