Hiro Ueno

Art director | Concept artist |
ComfyUIとクリエイティブ・エージェントで包括的なアート制作システムを構築中

[ComfyUI @ComfyUI]

Sora 2 API Node is now in Comfy!

Resolutions:
- Sora-2: 720 x 1280, 1280 x 720
- Sora-2 Pro: 720 x 1280, 1280 x 720, 1024 x 1792, 1792 x 1024

No watermarks!
Lengths: 4s, 8s, or 12s

[Trung Phan @TrungTPhan]

James Cameron says an opportunity with AI-generated videos is to make more blockbuster films by cutting “in half” the cost of computer-generated graphics.

Cost-saving isn’t about “laying off VFX staff” but about “doubling their speed to completion on a given shot.”

With AI integrated into VFX workflows, Cameron told Meta’s CTO Andrew Bosworth in April 2025 that the “cadence [can be] faster and your throughput cycle is faster and artists get to move on and do other cool things.”

Cameron joined the board of Stability AI to learn new generative AI tools and figure out how Hollywood can adopt the technology (he obviosuly previously pioneered VFX with The Abyss, Terminator, Avatar and Leo’s art book in Titanic).

[Hiro Ueno @hiro_ueno]

ComfyuiでNano Banana使っててなんか思ったよりAPI料金がかかると思ったら画像がけっこうトークン数使っていた。

ComfyのAPI使うと$0.03 / 1k トークンだけど画像入れると3ｋ以上になることが多いのでプロンプトの試行錯誤とかは別でやる方がいいのかも。

[Dria @driaforall]

LLMs are stateless.

We built Dria Mem Agent to change that: Making memory a first-class feature.

A 4B agent with local interoperable memory across Claude, ChatGPT and LM Studio. It turns LLMs from stateless chat into stateful agents with persistent human-readable memory.

[『アズワン／AS ONE』8.22 Fri ROADSHOW @asone_movie]

◤◢◤◢◤◢◤◢◤◢
『#アズワン／AS ONE』
主題歌入り予告映像
◤◢◤◢◤◢◤◢◤◢

地球の高校生・ヨウ “巡星”の少女・ラコ。
違う星、違う時に生きる二人。
運命に導かれた二人が出会ったとき、星の未来が変わるー

#白岩瑠姫（#JO1）
#白石晴香 #武内駿輔 #日笠陽子
#丸山隆平
主題歌：「巡星」 RUKI

監督：#静野孔文
キャラクターデザイン： #貞本義行
メカニックデザイン：#形部一平

[Hiro Ueno @hiro_ueno]

ComfyuiがNano BananaのAPIに対応。いろいろ早い・・・

[ComfyUI @ComfyUI]

ComfyUI now supports nano-banana via native API Nodes!

🔹 Supports both text-to-image & image-to-image generation
🔹 Impressive character consistency and fidelity
🔹 Text output capability
🔹 Customizable resolutions & aspect ratios

Getting Started
1. Update ComfyUI to the nightly ( Stable and Desktop will be ready soon)
2. Double-click canvas → Search "Google Gemini Image" → Run!

🧵👇

[ComfyUI @ComfyUI]

🍌 Nano Banana + Wan 2.2
🎤 Hosts: Purz & Fill
⏲️ August 29th - 6pm EST / 3pm PST
📍 Live on YouTube, X, and Twitch

Today we're going to use Nano Banana aka Gemini Flash Image to create start and end frames, then animate them with Wan 2.2!

[StudioGOONEYS(スタジオグーニーズ) @studio_gooneys]

#アズワン では、
「顔は作画、体やそれ以外はCG」 という、
前例のないハイブリッド手法にチャレンジしています！

CGと作画、それぞれが歩み寄り工夫を重ねたからこそ、 迫力と繊細さを兼ね備えた映像に仕上がりました！

アズワンならではの映像表現、ぜひ注目してみてください✨

#アズワン制作秘話 #アズワンCG班 #アズワン作画班

[Hiro Ueno @hiro_ueno]

本日公開の映画『アズワン/ ASONE』 に美術監督で参加させていただきました！ 是非劇場で観てください。
https://t.co/pTKVomRCTC

[Tencent Hy @TencentHunyuan]

We've heard the community! 📣📣📣

Following the open-source release of our Hunyuan 3D World Model 1.0, we're excited to introduce the new 1.0-Lite version, optimized for consumer-grade GPUs!
This is the first open-source, explorable world generation model compatible with CG pipelines, now more accessible than ever.

Key Technical Optimizations:
🔹Dynamic FP8 Quantization: We’ve cut VRAM requirements by 35%—from 26GB to under 17GB—making it easy to run on consumer GPUs without compromising performance.
🔹SageAttention Quantization: Our method quantizes the Q, K, and V matrices in the Transformer to INT8, combined with dynamic smoothing and hardware optimizations, to achieve an inference speedup of over 3x with less than 1% precision loss.
🔹Cache Algorithm Acceleration: By optimizing redundant time steps, we've significantly improved inference efficiency for a smoother user experience.

Now, developers can run a complex world model without the need for expensive, high-end hardware.

Try it now: https://t.co/swscD5KGu2
Github: https://t.co/vKq8ykY4Hu
Hugging Face: https://t.co/StC2E9kOmE
Technical Report: https://t.co/JmqbQ4iq9e

[toyxyz @toyxyz3]

https://t.co/C1QRAQo2PD

[Floor Eijkelboom (@ICML2026 🇰🇷) @FEijkelboom]

Flow Matching (FM) is one of the hottest ideas in generative AI - and it’s everywhere at #ICML2025.

But what is it? And why is it so elegant? 🤔

This thread is an animated, intuitive intro into (Variational) Flow Matching - no dense math required.

Let's dive in! 🧵👇

[Hiro Ueno @hiro_ueno]

なぜ90%の人がAIから期待した成果を得られないのか？

スタンフォード研究で判明した決定的な違い：「ツールとして使う」vs「パートナーとして協働する」。成功者はAIにフィードバックし、AIに質問させ、一緒に改善していく。

https://t.co/xsYXd2fQhJ

[Hiro Ueno @hiro_ueno]

『AIで仕事がなくなる』じゃなくて『AIで制作速度が数倍になる』が現実と思う。問題は効率化した分の時間を何に使うか。より多くのアイデア検討？品質向上？新規案件？選択肢は増える。

[Hiro Ueno @hiro_ueno]

すごそうだがこれもクローズド型なのね。

[Farhan @mhdfaran]

Bro… I stared at this for 10 minutes.

One is AI.
The other is also AI.
But only one feels human.

Midjourney V7 vs Higgsfield Soul.

What you’re about to see is insane. 🧵👇

[Hiro Ueno @hiro_ueno]

Intelの加速装置のようだ。

[ComfyUI @ComfyUI]

Exciting update for ComfyUI users on Intel hardware!

- The new OpenVINO custom node is now available for ComfyUI
- Unlock faster AI inference on Intel CPUs, GPUs, and NPUs
- No hardware upgrade needed, just install the ComfyUI-OpenVINO node

Enjoy smoother, more efficient workflows with Intel acceleration

🧵 ↓

[Hiro Ueno @hiro_ueno]

これ全部Veo3！すごい。

[Öner S. Biberkökü @OnerBiberkoku]

An official music video created entirely with Veo 3!

We collaborated with @Google to produce a music video for the queen of Turkish pop, affectionately known as the “Little Sparrow.”

On the 50th anniversary of her artistic career, we crafted this video using only Veo 3 and Imagen, in collaboration with Google. From the very beginning, our goal was not to make something that felt like an AI experiment, but to create a music video—one that truly fits the song. Instead of relying on eye-catching AI tricks, we aimed for the kind of emotional, feel-good, and hopeful story that the artist herself has always delivered throughout her career.

We’re deeply grateful to @Google for bringing us together with a living legend.

[septendec @septendec]

FLUX.1 [dev] Non-Commercial License の 2.d を比較（上から v1, v1.1, v1.1.1）。
v1→v1.1 の際に消えていた出力物の商用利用について v1.1.1 で復活しています。

v1: https://t.co/FVcVL2ksTP
v1.1: https://t.co/n3s8IDFlvl
v1.1.1: https://t.co/10aMOk0uBP

[Hiro Ueno @hiro_ueno]

ComfyUIでFlux編集機能が一通り揃った感じ。コンセプトワークのイテレーション速度が上がりそう

[ComfyUI @ComfyUI]

🎉Generate. Edit. Repeat. All locally.

Flux.1 Kontext [dev] is now open source and fully supported in ComfyUI!

Everything below is now enabled:

👉Multi-Step Editing with Context
👉Character Consistency
👉Local Editing
👉Style Reference
👉Object / Background Removal
👉Multiple Inputs
👉Text Editing

[ファイアーエムブレム ヒーローズ @FE_Heroes_JP]

【映像公開】
レギンさんのお供・ドローンが、アルフォンスさんたちのもとへ？
ドローンとの日常を描いたスペシャル映像をご覧ください♪

新英雄召喚イベントは本日16:00から開催！
「小人の国の後継者 レギン」が仲間になると、このドローンの「お供アクセサリー」が手に入りますよ♪
#FEヒーローズ