Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

DocJev is the fastest way to classify and split complex document packets ⚡️. (and the default is fully free and OSS!) I made a sick teaser video below. TY Opus 5.5 🙏 check it out:

Jerry Liu

85,991 subscribers

93,403 Aufrufe • vor 27 Tagen •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

25 Kommentare

Profilbild von Jerry Liu

Jerry Liuvor 27 Tagen

how i made the video 😂

Profilbild von Ricci Research

Ricci Researchvor 27 Tagen

351 bookmarks vs 301 likes is the most honest metric in dev tools, everyone plans to try it right after they finish the other 3,000 repos they bookmarked.

Profilbild von Asaf Mazuz

Asaf Mazuzvor 26 Tagen

Already on my directory 😄 Great tool Thank you for sharing it 🙏🏻

Profilbild von John Rood

John Roodvor 27 Tagen

Two invoices can have every page classified correctly and still get merged into one document. DocJev's default 0.1 boundary-review margin flags uncertain continuations too, not just cuts.

Profilbild von Victor Zhang

Victor Zhangvor 26 Tagen

can it classify between scanned pdf or text-based pdf?

Profilbild von Kaan Demirel

Kaan Demirelvor 27 Tagen

did you make the video with remotion? looks nice

Profilbild von Jerry Liu

Jerry Liuvor 27 Tagen

bro idk i pointed claude code opus 5.5 at it and made it go brrr

Profilbild von Fabian Schreiber

Fabian Schreibervor 26 Tagen

Really cool. If I understand correctly, you classify full documents and - currently - don't look into classifying chunks/subsections (e.g for RAG), right?

Profilbild von Gavin C.

Gavin C.vor 26 Tagen

This video is super cool! I'd love to know what tool you used to make it.

Profilbild von Roni Rechter

Roni Rechtervor 27 Tagen

Splitting document packets. The least glamorous demo and the one I want.

Profilbild von Shabah

Shabahvor 27 Tagen

this operate on the text layer tho right?

Profilbild von Jerry Liu

Jerry Liuvor 27 Tagen

yep. there are VLM options with llamaparse, which would increase parse latency (though not extract)

Profilbild von Shabah

Shabahvor 26 Tagen

Do you think it will be able to do page level classification of say a form vs a table? I have so many customers asking me how to chunk a table.

Profilbild von George Nick Gorzynski

George Nick Gorzynskivor 27 Tagen

I wonder how this would pair with PageIndex 👀

Profilbild von Gabriel Eisbruch

Gabriel Eisbruchvor 24 Tagen

For packet splitting, we'd validate page coverage separately: every source page belongs to exactly one output, with order preserved. That catches dropped or duplicated pages even when category predictions look sensible.

Profilbild von Lionel Ngendakuriyo

Lionel Ngendakuriyovor 26 Tagen

We are so done ❤️

Profilbild von aaron

aaronvor 27 Tagen

❌ doctor jev ✅ doc jev

Profilbild von zepa

zepavor 26 Tagen

So how? Like higgsfield

Profilbild von albs—

albs—vor 27 Tagen

really excellent video. is it remotion?

Profilbild von Brjan | AI Builder

Brjan | AI Buildervor 27 Tagen

DocJev sounds promising, but speed isn't everything, accuracy is key too

Profilbild von Vasant

Vasantvor 27 Tagen

I have captured multiple small size receipt into single photo. ( with the hope something similar solution will appear in future) Can it clarify and split ?

Profilbild von Jatin Garg

Jatin Gargvor 27 Tagen

packet splitting is where most doc systems fail silently. they either over-merge (lose structure) or over-split (create orphans). hard to benchmark.

Profilbild von ₮Ɽ₳Vł₴

₮Ɽ₳Vł₴vor 27 Tagen

dope!

Profilbild von Nichy

Nichyvor 27 Tagen

I need it to be able to distinguish between the layout providers for the same type of document class... for example, it needs to recognize that Layout X belongs to Supplier A and Layout Y belongs to Supplier B. Can it do that?

Profilbild von Vishal Sachar

Vishal Sacharvor 27 Tagen

Free OSS is a boon, but consider how it scales with volume. Performance gains must meet business needs.

Ähnliche Videos

Introducing DocJev - a lightning-fast OSS library for document classification and splitting with jev ⚡️ Give a document alongside some natural language category rules. Jev will predict the document category (classify) or the boundaries between sub-documents (split). It is 6x faster than gpt-5.6-luna with equivalent accuracy. You can choose between different OCR backends: liteparse: the fastest/most accurate free+OSS text parser out there. Perfect for digitalized documents. The speedups vs. luna above are inclusive of liteparse times. LlamaParse: the most advanced VLM-based solution for complex docs. Using this will of course add latency on document preprocessing. But it’s a good option if you want to cache the digitalized representations of documents for a variety of downstream tasks. Check it out: If you want to dive deeper into liteparse or LlamaParse check the links below: liteparse: LlamaParse:

Introducing DocJev - a lightning-fast OSS library for document classification and splitting with jev ⚡️ Give a document alongside some natural language category rules. Jev will predict the document category (classify) or the boundaries between sub-documents (split). It is 6x faster than gpt-5.6-luna with equivalent accuracy. You can choose between different OCR backends: liteparse: the fastest/most accurate free+OSS text parser out there. Perfect for digitalized documents. The speedups vs. luna above are inclusive of liteparse times. LlamaParse: the most advanced VLM-based solution for complex docs. Using this will of course add latency on document preprocessing. But it’s a good option if you want to cache the digitalized representations of documents for a variety of downstream tasks. Check it out: If you want to dive deeper into liteparse or LlamaParse check the links below: liteparse: LlamaParse:

Jerry Liu

217,617 Aufrufe • vor 29 Tagen

omg i had to test it myself and am stunned how good this is this product launch video was fully created by opus 5.5 with just one prompt i just told opus to go pick a SaaS and create a professional product launch video prompt below:

omg i had to test it myself and am stunned how good this is this product launch video was fully created by opus 5.5 with just one prompt i just told opus to go pick a SaaS and create a professional product launch video prompt below:

Moritz Kremb

96,242 Aufrufe • vor 25 Tagen

I gave GPT 6 SOL and OPUS 5.5 the same prompt. It was not even close. Opus 5.5 is THE BEST model we have ever seen. links to game below.

I gave GPT 6 SOL and OPUS 5.5 the same prompt. It was not even close. Opus 5.5 is THE BEST model we have ever seen. links to game below.

Max Blade

244,730 Aufrufe • vor 27 Tagen

Conditioning Control Panel v7 is here! Here is a little video I made to celebrate this launch , hope you enjoy, its a fever dream made with opus 5.5 and sleepless nights (you can also find the lore and lyrics down below)

Conditioning Control Panel v7 is here! Here is a little video I made to celebrate this launch , hope you enjoy, its a fever dream made with opus 5.5 and sleepless nights (you can also find the lore and lyrics down below)

CodeBambi

79,292 Aufrufe • vor 7 Tagen

Anthropic launched Sonnet 5.5, made it as good as Opus 5.5, then went ahead and nerfed Opus 5.5. Crazy marketing. We tested Opus 5.5 on launch day and again today, on the same open-world GTA 6 India prompt. Launch day vs today below. The difference is clearly visible.

Anthropic launched Sonnet 5.5, made it as good as Opus 5.5, then went ahead and nerfed Opus 5.5. Crazy marketing. We tested Opus 5.5 on launch day and again today, on the same open-world GTA 6 India prompt. Launch day vs today below. The difference is clearly visible.

Build Fast with AI

108,035 Aufrufe • vor 18 Tagen

I asked Opus 5.5 to picture a world in 2076 when AI had wiped out humanity. this is the story it told. animation, story, sound effects, and music all made by Claude. the video is coded in JavaScript.

I asked Opus 5.5 to picture a world in 2076 when AI had wiped out humanity. this is the story it told. animation, story, sound effects, and music all made by Claude. the video is coded in JavaScript.

ℏεsam

87,801 Aufrufe • vor 24 Tagen

We built the fastest PDF -> markdown parser in the world 🚀⚡️ AND it’s more accurate than any other open-source, model-free parser (pymupdf4llm, opendataloader, pdf-inspector, markitdown) on 3 standardized benchmarks: olmOCR0-bench, opendataloader-bench, ParseBench Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown. It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill. Check it out: Come check out LiteParse:

We built the fastest PDF -> markdown parser in the world 🚀⚡️ AND it’s more accurate than any other open-source, model-free parser (pymupdf4llm, opendataloader, pdf-inspector, markitdown) on 3 standardized benchmarks: olmOCR0-bench, opendataloader-bench, ParseBench Introducing LiteParse v2.1. The v2 base version was already the fastest document->text parser on the planet, and with this new release we’ve introduced markdown. It is fully open-source (Apache 2.0) and free, is usable from CLI/Rust/Node/Python/WASM, and is also installable as a one-click agent skill. Check it out: Come check out LiteParse:

Jerry Liu

364,389 Aufrufe • vor 4 Monaten

Opus 5.5 just one shot a trailer for my fishing game... Fully made in blender. $93 and 7 hours is all it took. Prompt in replies 👇 I also pushed an update that reworked a lot of stuff, check it out: :)

Opus 5.5 just one shot a trailer for my fishing game... Fully made in blender. $93 and 7 hours is all it took. Prompt in replies 👇 I also pushed an update that reworked a lot of stuff, check it out: :)

Blushney

43,805 Aufrufe • vor 10 Tagen

I asked Opus 5.5 to reverse engineer League of Legends And Dragonball FighterZ combine them and make a game out of it. This is the result...

I asked Opus 5.5 to reverse engineer League of Legends And Dragonball FighterZ combine them and make a game out of it. This is the result...

Aymane

896,982 Aufrufe • vor 16 Tagen

PART 2 I asked Opus 5.5 to reverse engineer League of Legends And Dragonball FighterZ combine them and make a game out of it. This is the result...

PART 2 I asked Opus 5.5 to reverse engineer League of Legends And Dragonball FighterZ combine them and make a game out of it. This is the result...

Aymane

72,237 Aufrufe • vor 15 Tagen

I asked Opus 5.5 to reverse engineer Dragonball FighterZ port it over to Threejs and make an MMO out of it. This is the result...

I asked Opus 5.5 to reverse engineer Dragonball FighterZ port it over to Threejs and make an MMO out of it. This is the result...

Aymane

40,478 Aufrufe • vor 16 Tagen

Today's video was a first look at Muse Wonder! Great to see Meta creating big models again, and Wonder is certainly within the same ballpark as other frontier models, although still quite a bit below Opus 5.5 in terms of performance Check it out!

Today's video was a first look at Muse Wonder! Great to see Meta creating big models again, and Wonder is certainly within the same ballpark as other frontier models, although still quite a bit below Opus 5.5 in terms of performance Check it out!

tonbi

22,337 Aufrufe • vor 6 Tagen

This is insane Opus 5.5 built THIS detailed 3d world with a lot of weird creative builds in one prompt and in less than 8h! Check it out here: (prompt is below)

This is insane Opus 5.5 built THIS detailed 3d world with a lot of weird creative builds in one prompt and in less than 8h! Check it out here: (prompt is below)

Leon Lin

29,783 Aufrufe • vor 25 Tagen

Claude Opus 5.5 completely MOGS GPT 6 Sol. Same prompt. Same task. Look at the results. Opus 5.5 has way better taste and design. It is not subtle. You can see it immediately. GPT 6 Sol is a step up from GPT 5.6 Sol. But next to Opus 5.5 it is not even close. Anthropic is back.

Claude Opus 5.5 completely MOGS GPT 6 Sol. Same prompt. Same task. Look at the results. Opus 5.5 has way better taste and design. It is not subtle. You can see it immediately. GPT 6 Sol is a step up from GPT 5.6 Sol. But next to Opus 5.5 it is not even close. Anthropic is back.

Bridgebench

55,327 Aufrufe • vor 27 Tagen

Something is wrong at OpenAI. Tibo and Sam said GPT-6.1 Sol would be faster today. Instead it appears to be running slower. Sonnet 5.5 and Opus 5.5 are much faster than Sol 6.1 and Grok 4.7 Check out the below comparison, same prompt on high reasoning with a 20 minute deadline.

Something is wrong at OpenAI. Tibo and Sam said GPT-6.1 Sol would be faster today. Instead it appears to be running slower. Sonnet 5.5 and Opus 5.5 are much faster than Sol 6.1 and Grok 4.7 Check out the below comparison, same prompt on high reasoning with a 20 minute deadline.

Eco

17,812 Aufrufe • vor 18 Tagen

I vibe coded this in a few hours using Codex OpenAI and also made the video is now live!! go check it out and drop your artwork below 🖼️

I vibe coded this in a few hours using Codex OpenAI and also made the video is now live!! go check it out and drop your artwork below 🖼️

Aarti

327,989 Aufrufe • vor 1 Monat

Fable 5.5 is INSANE... I asked it to make a 25 sec video about what the world will look like in 2076. this is what it made. weirdly it looks a LOT like Opus 5.5 lol.

Fable 5.5 is INSANE... I asked it to make a 25 sec video about what the world will look like in 2076. this is what it made. weirdly it looks a LOT like Opus 5.5 lol.

Layton Gott

39,932 Aufrufe • vor 17 Tagen

This Mac launch film is 100% code. Opus 5.5 made the notch, the widgets and the wallpapers from one brief. The bit that made the difference was the free apple-design skill. I had it tear apart my first version, list everything that looked cheap, and rebuild it. Full brief is here if you want to try it:

This Mac launch film is 100% code. Opus 5.5 made the notch, the widgets and the wallpapers from one brief. The bit that made the difference was the free apple-design skill. I had it tear apart my first version, list everything that looked cheap, and rebuild it. Full brief is here if you want to try it:

Charlie Hills

19,944 Aufrufe • vor 22 Tagen

this is certified f*cking gold for anyone paying for Claude Sonnet 5.5 costs half of what Opus 5.5 does on real office work across 44 jobs it scores 1844. Opus scores 1846 on coding it went from 10.3% to 70.6% in one version and passed Opus on the way most of what you pay Opus for, Sonnet 5.5 now does at half the price here's how to split the work: • give it to Sonnet 5.5 > bug fixes and well-scoped coding tasks > decks built from your own slide template > spreadsheets and polished documents > UI polish, it has a real eye for design > fast back-and-forth, it runs 30%+ faster • keep it for Opus 5.5 > long, open-ended work with judgment calls for hours > anything where one wrong assumption costs you a day • the setting that saves the most > Medium is the default in the Claude apps and Claude Code > on Low or Medium it beats Sonnet 5's best score for a tenth of the cost > raise effort only when the first answer falls short • the price > $2 in and $10 out, against $4 and $20 for Opus > cache reads $0.20 on both • check before you switch > if you ran Sonnet with thinking off, move to the new between_tools setting first whoever routes the weekly grunt work to Sonnet 5.5 and saves Opus for the hard calls ships more on half the budget, and that's the person who gets noticed

this is certified f*cking gold for anyone paying for Claude Sonnet 5.5 costs half of what Opus 5.5 does on real office work across 44 jobs it scores 1844. Opus scores 1846 on coding it went from 10.3% to 70.6% in one version and passed Opus on the way most of what you pay Opus for, Sonnet 5.5 now does at half the price here's how to split the work: • give it to Sonnet 5.5 > bug fixes and well-scoped coding tasks > decks built from your own slide template > spreadsheets and polished documents > UI polish, it has a real eye for design > fast back-and-forth, it runs 30%+ faster • keep it for Opus 5.5 > long, open-ended work with judgment calls for hours > anything where one wrong assumption costs you a day • the setting that saves the most > Medium is the default in the Claude apps and Claude Code > on Low or Medium it beats Sonnet 5's best score for a tenth of the cost > raise effort only when the first answer falls short • the price > $2 in and $10 out, against $4 and $20 for Opus > cache reads $0.20 on both • check before you switch > if you ran Sonnet with thinking off, move to the new between_tools setting first whoever routes the weekly grunt work to Sonnet 5.5 and saves Opus for the hard calls ships more on half the budget, and that's the person who gets noticed

Annatar.md

19,642 Aufrufe • vor 21 Tagen

I had early access to Opus 5.5 and it blew my expectations clear out of the water! Here are some things I made! Dark Souls

I had early access to Opus 5.5 and it blew my expectations clear out of the water! Here are some things I made! Dark Souls

Alex

606,859 Aufrufe • vor 27 Tagen