generative::media

What's actually shipping in generative AI. curated daily.

[generative::media @generativemedia]

openvlharness real world visual agentic tool-use with multimodal memory paper:

https://t.co/u87BC5cvBJ

[AK @_akhaliq]

OpenVLHarness

Real World Visual Agentic Tool-Use with Multimodal Memory

paper: https://t.co/DIsy4KDpMa

[generative::media @generativemedia]

i wrote about the exponential curve of the last 30 months of ai development: and i asked opus 5.5 to create the ultimate response to the test i use for all ai image generation models (make a picture

https://t.co/pHUTrxAwjh

[Ethan Mollick @emollick]

I wrote about the exponential curve of the last 30 months of AI development: https://t.co/lKh9THu9JH

And I asked Opus 5.5 to create the ultimate response to the test I use for all AI image generation models (make a picture of "an otter on a plane using wifi"), it made this video

[generative::media @generativemedia]

big boy rendered by computer

https://t.co/obbRGK8yHs

[Aravind Srinivas @AravSrinivas]

Big Boy rendered by Computer

[generative::media @generativemedia]

fibo 2.0 by @bria_ai_ is live on each::labs on day 0 bria’s new open-weights model for image generation and editing. generate images from text and pick your aspect ratio and output format. edi

https://t.co/AqlWQFVmHn

[each::labs @eachlabs]

FIBO 2.0 by @bria_ai_ is live on each::labs on day 0 🌈

Bria’s new open-weights model for image generation and editing.

→ Generate images from text and pick your aspect ratio and output format.
→ Edit photos with text instructions or precise masks.

Try it now ⤵

[generative::media @generativemedia]

paying full price for every bad take? seedance 2.5 draft completion is live on each::labs try your ideas as quick, low-cost 480p drafts. pick the take you love and finish it in 1080p. same m

https://t.co/7XxsDyU8Rb

[each::labs @eachlabs]

Paying full price for every bad take?

Seedance 2.5 Draft Completion is live on each::labs 🎬

→ Try your ideas as quick, low-cost 480p drafts.
→ Pick the take you love and finish it in 1080p.
→ Same motion and timing, in final-cut quality.

Try it now ⤵

[generative::media @generativemedia]

gemini 4 argon built a 3d periodic table synced to tom lehrer’s song, “the elements.” created in @antigravity, it animates orbiting electrons, atomic data, and light spectra.

https://t.co/ZT9eSb4uQ0

[Google DeepMind @GoogleDeepMind]

Gemini 4 Argon built a 3D periodic table synced to Tom Lehrer’s song, “The Elements.”

Created in @Antigravity, it animates orbiting electrons, atomic data, and light spectra.

[generative::media @generativemedia]

davinci resolve studio just got a very rare 20% discount

https://t.co/QfNhGp0yCt

[generative::media @generativemedia]

inception just launched mercury 3, which it calls the strongest diffusion language model yet. it’s 40x bigger than mercury 1 but still puts out 1,000+ tokens a second. the comparison is to the fast,

https://t.co/2bXKB4K8Ic

[The Rundown AI @TheRundownAI]

Inception just launched Mercury 3, which it calls the strongest diffusion language model yet.

It’s 40x bigger than Mercury 1 but still puts out 1,000+ tokens a second.

The comparison is to the fast, cheap tier, with Inception saying it matches or beats OpenAI’s GPT-6 Luna and Anthropic’s Claude Haiku 5.5.

Where ChatGPT and Claude write one word at a time, diffusion models draft many words at once and refine them, making them a strong fit for use cases like voice agents, live tutoring, and customer support, where any lag shows.

[generative::media @generativemedia]

claude can make animations now. gorgeous ones. then you sit there going “okay, cool… now what?” now you send it to descript and add a voiceover, music, captions and stock footage. claude for the mot

https://t.co/FUtDc52b8y

[Descript @descript]

Claude can make animations now. Gorgeous ones. Then you sit there going “okay, cool… now what?”

Now you send it to Descript and add a voiceover, music, captions and stock footage.

Claude for the motion, Descript for the rest.

https://t.co/cloFnxmYlF

[generative::media @generativemedia]

from oct 10 to 13, kling ai joined acfm 2026 for deep exchanges with the global film industry! focus on kling 4.0’s model capabilities, we brought together leading filmmakers across asia to explore t

https://t.co/YDnSWxlpo4

[Kling AI @Kling_ai]

From Oct 10 to 13, Kling AI joined ACFM 2026 for deep exchanges with the global film industry!

Focus on Kling 4.0’s model capabilities, we brought together leading filmmakers across Asia to explore the future of story-driven AI and its transformative impact on professional filmmaking.

Kling AI continues to pioneer AI video integration in professional production. By collaborating with global creators, we’re bringing generative AI into real-world workflows to turn creative vision into cinematic reality.

[generative::media @generativemedia]

join us to celebrate the launch of krea 3, our new frontier model for image editing. we’ll have drinks, food, live demos, and a dj set throughout the night. rsvp below

https://t.co/c0MCjXO57X

[Krea @krea_ai]

join us to celebrate the launch of Krea 3, our new frontier model for image editing.

we’ll have drinks, food, live demos, and a DJ set throughout the night.

rsvp below 👇

[generative::media @generativemedia]

minimax m3.1 flash shows china's ai price race has no floor in sight

https://t.co/wKnSLdpsdT

[generative::media @generativemedia]

runway mcp is now in grok bot. bring state-of-the-art generation into your chat and create anything you can imagine, without leaving grok. get started at the link below.

https://t.co/55FYkQnrcC

[Runway @runwayml]

Runway MCP is now in Grok Bot.

Bring state-of-the-art generation into your chat and create anything you can imagine, without leaving Grok. Get started at the link below.

[generative::media @generativemedia]

painting & image editing workspace with live brush stroke previews: meet moko

https://t.co/m66oZ6VPTi

[generative::media @generativemedia]

kling ai drops version 4.0 with stereo audio, multilingual lip sync, and a lighter flash variant

https://t.co/53yPEMuhnH

[generative::media @generativemedia]

film the dance once and motion transfer teaches it to any character, right from chatgpt. no dancing skills required.

https://t.co/681CSBxKNB

[Pika @pika_labs]

Film the dance once and Motion Transfer teaches it to any character, right from ChatGPT. No dancing skills required.

[generative::media @generativemedia]

a 3b-active model can now write long-term memory as well as claude opus 5. tencent hunyuan trained hy-memory-30b-a3b for one job. it decides what an agent stores, how precisely, and how new evidence

https://t.co/0GJaT5lnXx

[DAIR.AI @dair_ai]

A 3B-active model can now write long-term memory as well as Claude Opus 5.

Tencent Hunyuan trained Hy-Memory-30B-A3B for one job. It decides what an agent stores, how precisely, and how new evidence updates what is already there.

The same model runs inside Mem0, MemOS, and Hy-Memory. On fixed 20% benchmark subsamples, it ranks first on all four evaluated harnesses, including held-out EverMemOS, with a 55.49% macro average against 54.62% for Opus 5.

Training has two stages. On-policy distillation teaches each framework's write protocol. RL then rewards the writer when later questions can be answered from what it stored, with a penalty for storing too much.

https://t.co/If1TULtR0y

Chat with Paper: https://t.co/0GxMJwHTFn

[generative::media @generativemedia]

ai video is evolving. where does the industry go from here? at advertising week new york 2026, kling joined wpp and adobe to discuss how ai is reshaping creative workflows and unlocking new possibi

https://t.co/eG29VFSRVO

[Kling AI @Kling_ai]

AI video is evolving. Where does the industry go from here? 👀

At Advertising Week New York 2026, Kling joined WPP and Adobe to discuss how AI is reshaping creative workflows and unlocking new possibilities for advertising and brand production.

Watch the highlights.

[generative::media @generativemedia]

going big with kling 4.0 #kling4 #klingai

https://t.co/ggaZcKgtmG

[Kling AI @Kling_ai]

Going big with Kling 4.0 👀 #Kling4 #KlingAI

[generative::media @generativemedia]

meet voxel51 3.0: a unified multimodal data platform for physical ai

https://t.co/kD9blR39kE