Photogenic Weekend (@PhotogenicWeekE) profile picture

Photogenic Weekend

@PhotogenicWeekE

PC Watch記事 https://teapawt.net/t.co/J5HjvfGIeM テクノエッジ記事(9月から再開！セミナーは10月から) https://teapawt.net/t.co/efoNfGUFRH

 東京 渋谷区
Joined September 2011
  • Tweets15.6K
  • Following974
  • Followers11.2K
3.1KMedia
Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)Photo by Photogenic Weekend (@PhotogenicWeekE)
無料の画像生成AI「Moss-Image」がすごい　日本語も漫画も生成、商用利用OKの実力を試した 画像だけでなくテキスト表示など複雑な表現に対応した画像AIがオープンウェイトモデルかつ、自由度が高いラインセンスで登場しました。中国の行政からの大きな投資を受けています。 ascii.jp/elem/000/004/445/44…
無料の画像生成AI「Moss-Image」がすごい 日本語も漫画も生成、商用利用OKの実力を試した (1/4)

無料の画像生成AI「Moss-Image」がすごい　日本語も漫画も生成、商用利用OKの実力を試した (1/4)

日本語描画や多彩な画風に対応した、オープンウエイトの画像生成AI「Moss-Image」を試す。行政の支援を受ける開発体制と事業戦略、今後の普及や画像生成AI市場への影響を考える。

ascii.jp
2
39
9
270
37,740
Photogenic Weekend retweeted
VIDEO REBIRTH (@video_rebirth) profile picture
VIDEO REBIRTH
@video_rebirthOct 25
We're open-sourcing HyperFlow Infinity, the second release in the HyperFlow series. HyperFlow Infinity achieves unlimited-length video generation: a minimal agent harness where HyperFlow generates a shot, watches it, updates a memory, and directs the next—in a recursive loop. Our first film, "Alice in Wonderland: An Endless Wander", runs 60 minutes in a single pass, generated by Hyperflow Infinity, without extra model. Code: github.com/Video-Rebirth/hyp…
25
33
9
367
94,202
Photogenic Weekend retweeted
Niko Veit (@coldniko) profile picture
Niko Veit
@coldnikoOct 24
Deepseek-V4.1-Flash now runs on 64GB of RAM with Strata. Using GSQ-RCO quantizations by @mhelcig
145
311
100
3,143
289,002
Photogenic Weekend retweeted
Mia (@MiaAI_lab) profile picture
Mia
@MiaAI_labOct 22
MiniMax-M3.1-Flash open weights drop tomorrow.
Image from Mia (@MiaAI_lab)
16
13
256
12,148
Photogenic Weekend retweeted
Wildminder (@wildmindai) profile picture
Wildminder
@wildmindaiOct 20
ComfyUI now supports color‑depth control for MOSS‑image. Sweet. github.com/Comfy-Org/ComfyUI…
Image from Wildminder (@wildmindai)
Image from Wildminder (@wildmindai)
4
24
2
266
15,482
Photogenic Weekend retweeted
Rebel AI (@realrebelai) profile picture
Rebel AI@realrebelaiOct 20
MOSS-Imagine an Grok Imagine adaptation based finetune of MOSS-Image-SFT huggingface.co/realrebelai/M…
Image from Rebel AI (@realrebelai)
1
2
47
2,166
Photogenic Weekend retweeted
Kohya Tech (@kohya_tech) profile picture
Kohya Tech
@kohya_techOct 20
MOSS-Image用の画像キャプション付け（long/medium/shortの3バージョン）をVLMで行うスクリプトをClaudeに書いてもらった。sd-scriptsで使える形式で出力されます。 gist.github.com/kohya-ss/6bd… Gemma4やQwen3.5/3.8系で動かしてください。JSONレイアウト出力は未対応です。
OpenAI互換APIサーバのVLMを呼び出し、MOSS-Image/sd-scripts向けの画像キャプションを作成する

OpenAI互換APIサーバのVLMを呼び出し、MOSS-Image/sd-scripts向けの画像キャプションを作成する

OpenAI互換APIサーバのVLMを呼び出し、MOSS-Image/sd-scripts向けの画像キャプションを作成する - caption_images_moss.py

gist.github.com
10
1
63
4,531
Photogenic Weekend retweeted
Rebel AI (@realrebelai) profile picture
Rebel AI@realrebelaiOct 19
RebelReal LoRA (MOSS-Image) civitai.com/models/2181922/r… huggingface.co/realrebelai/R…
Image from Rebel AI (@realrebelai)
Image from Rebel AI (@realrebelai)
Image from Rebel AI (@realrebelai)
1
1
6
903
Photogenic Weekend retweeted
Plat 🖼️ (@p1atdev_art) profile picture
Plat 🖼️@p1atdev_artOct 19
モデルとカスタムノード公開した huggingface.co/p1atdev/MOSS-…
p1atdev/MOSS-Image-SFT-KV-Cache-LoRA · Hugging Face

p1atdev/MOSS-Image-SFT-KV-Cache-LoRA · Hugging Face

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
Plat 🖼️ (@p1atdev_art) profile picturePlat 🖼️@p1atdev_artOct 19
MOSS Image で KV キャッシュできるようにするカスタムノードチャッピーが作ってくれた
Image from Plat 🖼️ (@p1atdev_art)
2
5
1
55
4,114
Photogenic Weekend retweeted
ゆずき (@uzuki425) profile picture
ゆずき
@uzuki425Oct 18
FLUX 3 Image Kleinプロンプトの書き方｜めぐチャンネル @uzuki425 note.com/ai_meg/n/n34f2423c3…
FLUX 3 Image Kleinプロンプトの書き方｜めぐチャンネル

FLUX 3 Image Kleinプロンプトの書き方｜めぐチャンネル

FLUX 3 Image Klein Reference Studio で、どう書くと何が出てくるかをまとめたガイドです。前半でプロンプトの仕組みと書き方を、 後半で機能ごとの実際の依頼文・キャプション・生成画像を紹介します。仕組みの詳細はテクニカルガイド、 API は API リファレンス を参照してください。 サンプルの条件: INT8、16 ステップ、タイムステップのシフト 16、seed...

note.com
4
1
12
2,339
Photogenic Weekend retweeted
ITOK (@ITOKtw) profile picture
ITOK@ITOKtwOct 18
Flux3 klein の改良テンプレート，解像度が4MPで，Timestep Shiftが入っています。良くなってはいますが，指は時々怪しくなります。 ・baseの方が安定 ・prompt upsamplrを使わないシンプルなプロンプトは変形しやすい ・寝転んだりする画像では，アスペクト比を合わせた方が良い（1:1は失敗しやすい）
Image from ITOK (@ITOKtw)
Image from ITOK (@ITOKtw)
Image from ITOK (@ITOKtw)
Image from ITOK (@ITOKtw)
ITOK (@ITOKtw) profile pictureITOK@ITOKtwOct 16
Flux3 t2i，解像度が低いためかもということでint8で2MP。それでも1/4ぐらいで出ます。背景はいい感じなだけに残念です。
Image from ITOK (@ITOKtw)
Image from ITOK (@ITOKtw)
1
1
1
15
2,329
Photogenic Weekend retweeted
buchi_ (@mix_buchi_) profile picture
buchi_
@mix_buchi_Oct 18
MOSS-Imageのワークフロー一気に来てます #ComfyUI
Image from buchi_ (@mix_buchi_)
1
26
1,889
Photogenic Weekend retweeted
AI画社(AI Image Studio) (@seki_images) profile picture
AI画社(AI Image Studio)
@seki_imagesOct 17
Radiance Chrome Voluptuous for Flux.3 Klein v1.0 has been released on CivitAI and Hugging Face (URLs in the comments). Key features: - Reversal film-inspired photographic style, with strong contrast and rich, saturated colors. - Significantly reduces the overly grayish tones and excessively fine-grained noise/artifacts of the base model. - Improved lighting and shadow rendering, especially exceptionally powerful and expressive sunlight. - Based on my personal aesthetic preferences, female characters tend toward larger breasts, slim waists, and wider hips. NSFW supoorted. Recommended strength: 1.0 Recommended sampler and scheduler: euler_simple
Image from AI画社(AI Image Studio) (@seki_images)
Image from AI画社(AI Image Studio) (@seki_images)
Image from AI画社(AI Image Studio) (@seki_images)
Image from AI画社(AI Image Studio) (@seki_images)
3
3
29
2,626
Photogenic Weekend retweeted
MOSS (@MOSS_Global) profile picture
MOSS
@MOSS_GlobalOct 17
👀 want EARLY ACCESS to our next models? join the MOSS Discord for upcoming early access opportunities, help with MOSS-Image, and ideas from fellow creators. love seeing the community's LoRAs already 🫶 come show us yours join MOSS discord 👉 discord.gg/z7TD5ehYts
Image from MOSS (@MOSS_Global)
18
12
1
77
5,411
Photogenic Weekend retweeted
Qwen Developers (@QwenDevs) profile picture
Qwen Developers
@QwenDevsOct 16
We're co-hosting Open Together with @huggingface this Friday in San Francisco! Doors open at 6PM, RSVP: luma.com/OpenTogether See you guys there!
35
40
18
566
63,198
Photogenic Weekend retweeted
Kohya Tech (@kohya_tech) profile picture
Kohya Tech
@kohya_techOct 17
MOSS-Imageから対応するぜ～(･∀･)
1
4
1
62
3,184
Photogenic Weekend retweeted
Kohya Tech (@kohya_tech) profile picture
Kohya Tech
@kohya_techOct 17
RTX Spark発売された～(･∀･)　 sd-scripts、Musubi TunerはRTX SparkでのLoRA学習に対応済みなので、手に入れた方はぜひお試しください。動作報告など、お待ちしています。
3
18
2,159
Photogenic Weekend retweeted
MOSS (@MOSS_Global) profile picture
MOSS
@MOSS_GlobalOct 16
🤗 MOSS-Image is now open source on @huggingface . Built on a 5B Diffusion Transformer (DiT), MOSS-Image brings Chinese & English text rendering and controllable layouts to image generation — from posters and infographics to creative visuals.@Open_MOSS What it offers: 📝 Long prompts & text rendering — up to 4K-token prompts, including multi-line text 📐 Experimental JSON layout control — specify text, objects, and their positions using bounding boxes 🖼️ Flexible generation — multiple aspect ratios, resolutions, and customizable inference settings 🛠️ Developer workflows — Python (Diffusers), Gradio, ComfyUI, and a Prompt Enhancer Skill 📦 Base & SFT checkpoints are both available, with open model weights and Apache-2.0 licensed inference code. 🙌 Huge thanks to @sgl_project @vllm_project @ModelScope2022 for the Day-0 support!
Image from MOSS (@MOSS_Global)
35
62
32
523
65,955
Photogenic Weekend retweeted
Ostris (@ostrisai) profile picture
Ostris
@ostrisaiOct 16
So @Open_MOSS MOSS Image learns image editing fast. It picked up my "Make this person a cyclops" concept almost immediatly. I added support to train it as an edit model in AI Toolkit. I am training a style reference LoRA now.
Image from Ostris (@ostrisai)
Image from Ostris (@ostrisai)
9
11
2
264
18,156
Photogenic Weekend retweeted
まゆひらa (@riddi0908) profile picture
まゆひらa
@riddi0908Oct 15
同意したらOK black-forest-labs/flux-3-image-klein-comfy huggingface.co/black-forest-…
black-forest-labs/flux-3-image-klein-comfy · Hugging Face

black-forest-labs/flux-3-image-klein-comfy · Hugging Face

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
1
7
60
4,080
Load more