Which Claude model should you build with?
Our new guide breaks down when to use Fable 5.2, Opus 5.5, Sonnet or Haiku. It also covers when to adjust effort instead of switching models, and how to save cost by putting subagents on smaller models.
https://t.co/1VmXqhkZB0
A cinematic documentary film about Al Capone created by Opus 5.5. 🔥
Just a single prompt + ref images.
What do you guys think of the film?
I’ll drop the sample prompt in the comments.
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.
They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
🚨 "مطوّرون صينيون يوجهون صفعة قوية لصناعة فيديوهات الذكاء الاصطناعي المدفوعة!".. أداة واحدة تنهي عصر الاستوديوهات والكاميرات!
مشروع مجاني ومفتوح المصدر على GitHub يُمكّنك من تحويل أي صورة ثابتة مع ملف صوتي إلى فيديو ناطق متزامن بدقة عالية يستمر لعدة دقائق متواصلة: LongCat Avatar!
🔥 ما الذي تقدمه الأداة فوراً؟
• رفع صورة واحدة + ملف صوتي = فيديو كامل: مزامنة شفاه وحركات وجه فائقة الدقة لعدة دقائق دون تقطع.
• تدمير تكاليف الاستوديوهات: ما كان يتطلب كاميرات، إضاءة، واستوديوهات مونتاج باهظة أصبح يخرج من مجرد Repository واحد.
• مفتوح المصدر بالكامل: لا اشتراكات شهرية، ولا نقاط تستهلكها في المنصات الاحتكارية.
صناعة الفيديوهات الرمزية (Avatars) لم تعد احتكاراً للشركات الكبرى.. الاستوديو بالكامل بات يعمل على حاسوبك محلياً بضغط زر!
📌 احفظ التغريدة بمفضلتك فوراً قبل أن تفرُض عليها القيود، ورابط الـ Repo تجده في أول تعليق 🔖👇
Introducing AI editing for Premiere Pro.
Give one prompt and Motion edits the ENTIRE video. Then open it in Premiere as a real project.
Every cut, every caption, every keyframe: editable.
Stop editing by hand like it’s 2020.
История, конечно, видела всякое варварство, но такого чудовищного акта геноцида человечество еще не знало. За какую-то неделю бесчеловечные бандеровцы сожгли 3/4 американцев, активно поддерживавших в интернете Трампа, 3/4 французов, активно поддерживавших в интернете Ле Пен, 3/4 немцев, активно поддерживавших в интернете АдГ, 3/4 итальянцев, активно поддерживавших в интернете Ваначчи, 3/4 поляков, активно продвигавших в интернете антибандеровскую повестку... Не менее 200 миллионов жертв. И никто даже не заплакал. Ужасный, ужасный мир.
Compositor (v1.4) is out! This update brings big GPU performance improvements throughout the app. Frames render 10x faster, it handles large files with ease, and everything feels buttery smooth. And the app is now just 11MB.
https://t.co/QQu47IK66y
🚨 وداعاً ElevenLabs للأبد: مطور بمفرده يطرح بديلاً مفتوح المصدر ومجانياً 100% يعمل بالكامل على جهازك الشخصي! 🎙️⚡️🔓
المشروع ينفجر على GitHub ويتخطى 19.4 ألف نجمة (Stars).. والقدرات تتفوق بشكل اكتساحي على الشركات المليارية:
ما الميزات الثورية التي يقدمها؟
🎯 استنساخ أي صوت بدقة خرافية من عينة صوتية قصيرة ونقية!
🌍 دبلجة أي مقطع فيديو إلى 646 لغة مختلفة! (مقارنة بـ ElevenLabs الذي يدعم 32 لغة فقط)!
📚 إنشاء كتب صوتية، إملاء، وتحويل الصوت إلى نصوص (Transcription) بثوانٍ!
⚙️ حرية اختيار تنقل بين 14 محرك توليد صوتي (TTS Engines) مختلف وليس محركاً واحداً!
الأهم من كل هذا؟
❌ لا اشتراكات شهريّة.
❌ لا دفع بناءً على عدد الحروف.
❌ لا حدود للاستخدام، ولا توجد ذرة ملف صوتي تخرج خارج جهازك (خصوصية مطلقة 100%)!
عصر الـ Local AI يكتسح مجدداً ويجعل الأدوات المدفوعة في خبر كان..
احفظ التغريدة فوراً لتعود إليها، رابط المستودع البرمجي بالأسفل! 🧵👇
ESTO ES JODIDAMENTE INCREÍBLE
un dev en solitario acaba de abrir el código de un reemplazo 100% GRATIS de ElevenLabs que se ejecuta completamente en tu propia máquina.
el repo de GitHub está en 19.4K estrellas.
te permite:
→ clonar una voz desde un clip de referencia limpio
→ doblar cualquier video a 646 idiomas
→ generar audiolibros, dictado, transcripción
→ elegir entre 14 motores TTS en lugar de uno
ElevenLabs soporta 32 idiomas. esto hace 646.
sin cobro por carácter. sin límites de uso. ningún audio sale nunca de tu computadora.
guarda esto para después.
os dejo el repo abajo
EYES WIDE SHUT: THE MISSING 20 MINUTES
You know, this is as good a time as any to finally address the missing 20 minutes from Eyes Wide Shut. It’s a story that has been circulating for many years, so let me tell you what I actually know and don’t know.
For context, my father always had his outtakes incinerated once a film was finished, specifically to prevent the studio from later re-editing his work or creating alternative versions.
There has been a lot of speculation that some 20 minutes were removed from Eyes Wide Shut after he died. This will be very disappointing to many people, but I certainly did not hear anything about a missing 20 minutes at the time, nor have I heard anything since that would confirm such a section ever existed. I guess you could stop reading right here. But if you’re interested in the details, please read on.
Of course, I was not in a position to know everything that may have happened before or immediately after my father’s death. At the time, I was living in Los Angeles, writing the score for a different movie. I flew to England, utterly traumatized, to be there for the preparations for my father’s funeral and burial. But I had to return to LA the day after his burial—stricken with grief and exhaustion—to complete the score I was contractually obligated to finish, which I barely managed. So I wasn’t there, in more ways than one, for whatever discussions may have followed.
The only alteration I heard about at the time was the use of CGI—computer-generated figures—to obscure some of the nudity in the orgy scene, which Warner Brothers wanted in order to secure an R rating for the film. But I did not hear that any shots or scenes were removed.
When my father died, Eyes Wide Shut had reached picture lock and was moving through the final stages of post-production. It had just been shown to the studio heads, I assume with a temporary music track, while the final sound mix and some of the music work were still to be completed.
This was 1999, in the era of physical film distribution. The film was edited on Avid, but that should not be confused with the kind of permanent digital archive people imagine today. The Avid material consisted of standard-definition video transfers made from the rushes—720 × 576 pixels—and used for editing. They were working copies, not a high-resolution digital preservation of the original negative. There wasn’t a high-resolution digital version of every take sitting on a server somewhere, as people might reasonably imagine today.
There were, of course, all the 35mm rush prints, as well as the original camera negative containing the outtakes. Eventually, long after the film was finished and released, the 35mm rush prints and the original negative of outtakes were incinerated. The original master negative of the finished film and the intermediate film elements made from it were, of course, kept in storage.
Once the picture was locked, the original camera negative would have been physically cut to match the final Avid edit. Once the negative itself had been cut, you couldn’t simply undo those cuts and rearrange the film: each cut destroyed a frame at both the head and tail of the shot. The master negative of the finished film was then used to make interpositives, from which internegatives were made, and those internegatives would have been used to make the 35mm theatrical release prints. The original negative of the outtakes never went through that process because, of course, those shots were not part of the finished film.
Someone might say, “But then, if the missing 20-minute edit exists, it would be preserved on the Avid!” But editing the film on Avid does not mean all of that material was preserved indefinitely. This was not an era in which every piece of editorial media was automatically retained on servers for decades, and I have no knowledge of whether any of the Avid editorial material from the production still exists. That’s a question for the …
CONT. Thread👇
DAVINCI RESOLVE ONLINE SEMINARS.
These six free 90 minute online seminars are ideal for filmmakers, editors, colorists and VFX artists, with something for both new and experienced users of DaVinci Resolve.
https://t.co/wOVQ54qrNo
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.
Sneaky Spielberg hid a cut during the "fence oner" in DISCLOSURE DAY.
The car’s passenger door is CLOSED on the approach.
The camera wipes across the back of the car and he cuts to another take.
The passenger door is now OPEN, allowing the camera to enter the car.
WIPE SWAP!