Top Tweets for #diffusiongemma
#DiffusionGemma is not for consumer hardware, yet.
Q4_K_M can get you only ~16k context size.
Q6 tanks at 8080:
Rather than generating text word by word, Google's experimental open-source model drafts entire passages simultaneously using diffusion, resulting in up to 4x faster inference.
#AI #ArtificialIntelligence #DiffusionGemma #GenerativeAI #data
https://t.co/wvvTyA5B27
@UnslothAI Fast inference is exciting — but what you prompt it with still determines the output quality. ⚡
Save your best DiffusionGemma prompts and never lose them at https://t.co/cK4Ih3QrEF — free prompt management for AI power users. 🚀
#DiffusionGemma #UnslothAI #PromptEngineering
Google's DiffusionGemma doesn't type — it prints. 256-token blocks generated in parallel, up to 4x faster on GPUs, 1000+ tok/s on an H100, Apache 2.0 open weights. The trade-off and where to get it — in one cheatsheet. #DiffusionGemma #AI #LLM
拡散言語モデルをMatrix風推論デザインにしてみた。気持ち良い
#diffusiongemma
Tu savais @Google a lancé DiffusionGemma ? Un modèle ouvert qui imprime le texte par blocs de 256 mots, jusqu'à 4x plus vite en local. La presse d'imprimerie contre la machine à écrire. #DiffusionGemma #IA https://t.co/wSyZ3UBFWK
📢[6/11]IMの気になるニュース
#Google 、 #拡散型テキスト生成モデル 「 #DiffusionGemma 」公開 ローカルGPUで毎秒1000トークン超💻🤖⚡
「1トークンずつ文字を紡ぐ従来のAIの常識を覆し、256文字のブロックを一括で並列生成して自己修正する超高速な仕組みをご存知ですか？」
米Googleは、画像生成でおなじみの「拡散（diffusion）」手法をテキスト生成に応用した実験的AIモデル「DiffusionGemma」を公開した。
従来の左から右へ1文字ずつ出力する自己回帰型モデルとは異なり、並列処理によってデコードのボトルネックを解消し、コンシューマー向けの「GeForce RTX 5090」でも毎秒700トークン以上、NVIDIA「H100」では毎秒1000トークンを超える驚異的な爆速推論を実現している。
開発者や研究者は、クラウドの通信や文字が1つずつ出力されるのを待つ旧来の遅延フェーズから完全に脱却し、ローカル環境での圧倒的な超高速処理を活かしてインライン編集やコード補完といったリアルタイム性の高い次世代AIシステムを構築していくことが、開発効率を極限まで引き上げるための絶対的なカギとなる。
https://t.co/CQ8vqrBwnP
@itmedia_news
DiffusionGemma uses text diffusion to generate 256 tokens in parallel, unlocking ultra-fast local AI workflows.
Do you think diffusion models are the future of AI generation?
Share your opinion in the comments! 👇
#GoogleAI #DiffusionGemma #ArtificialIntelligence
一句话总结：256K 在这个扩散模型 #diffusiongemma 上是跑不起来的，不是你配置的问题，是扩散架构在 24GB 显存上的物理极限。你想要长上下文，自回归的 Gemma 4 方案才是正解。
Meet DiffusionGemma ⚡ Our latest experimental open model (Apache 2.0) that generates text up to 4x faster.
Instead of predicting and typing just one word at a time like most language models, it drafts and refines entire blocks of text simultaneously.
Here’s how it works 🧵 ↓
خدا لعنتتون کنه!
بیش از 1 ترابایت پاک کردم و حدود 500 گیک گلچین کردم، برای روزی که اینترنت قطع میشه و از مدلهای آنلاین نمیشه استفاده کرد.
امشب هم #diffusiongemma رو دانلود میکنم.
https://t.co/FXN7ROtFQS
Google has introduced DiffusionGemma to speed local AI output through parallel text diffusion, but lower quality than Gemma 4 keeps trade-offs visible.
#AI #DiffusionGemma #TextDiffusion #Google #GoogleAI #AIModels #OpenSourceAI #OnDeviceAI #AIResearch
👀 DiffusionGemma: Google lancia un nuovo modello open sourceper esecuzione in locale che elabora 256 token in parallelo, usa attention bidirezionale e si auto-corregge in tempo reale.
https://t.co/WXcj0wvTva
#DiffusionGemma #geminidiffusion #google #news
DiffusionGemma, il nuovo modello open source di Google
https://t.co/FQ3ZsjTNXc
#DiffusionGemma #geminidiffusion #google #news
Spent some time reading about #DiffusionGemma today.
The most interesting takeaway wasn't the benchmarks—it's that Google is exploring a completely different direction for language models instead of simply making them larger.
Worth watching.
#AI #LLM #GoogleAI #MachineLearning
Google تكشف DiffusionGemma: طريقة جديدة قد تجعل ردود الذكاء الاصطناعي أسرع بكثير
#الذكاء_الاصطناعي #Google #DiffusionGemma #انتشار_النص #النماذج_اللغوية #Gemma https://t.co/In7cAyXHWy
🤖🚀 Google has launched DiffusionGemma, an open AI model that generates text in parallel, paving the way for faster and more efficient AI systems.
Read the full blog here 👇️
https://t.co/37IFanGhWV
#Google #DiffusionGemma #AI #MachineLearning #OpenSourceAI #TechNews
Google 發表實驗性開放模型 DiffusionGemma！透過擴散方式而非逐字形式生成文字，在單一 GPU 單一使用者模式運行時速度最高可比傳統語言模型快 4 倍。 #Google #DiffusionGemma #AI模型 https://t.co/q1PWUqRJIS
Google’dan DiffusionGemma geldi 👀
Diffusion mimarisiyle text generation’da devrim niteliğinde bir yaklaşım: 256 token’ı paralel üretip aynı anda düzeltiyor. 26B MoE (sadece 3.8B aktif parametre)
Tamamen açık kaynak (Apache 2.0).
#AI #Gemma #DiffusionGemma #OpenSource
Meet DiffusionGemma!
An experimental open model that explores a fast approach to text generation, released under an Apache 2.0 license.
Moving beyond sequential, token-by-token processes to generate entire blocks of text simultaneously. Here’s what’s new with DiffusionGemma: 👇
Most Popular Users
Elon Musk 
@elonmusk
240.2M followers
Barack Obama
@barackobama
119.3M followers
Donald J. Trump
@realdonaldtrump
111.6M followers
Cristiano Ronaldo
@cristiano
109.4M followers
Narendra Modi
@narendramodi
106.9M followers
Rihanna
@rihanna
97.4M followers
NASA
@nasa
92.1M followers
Justin Bieber
@justinbieber
90.7M followers
KATY PERRY
@katyperry
87M followers
Taylor Swift
@taylorswift13
80.9M followers
Lady Gaga
@ladygaga
72.4M followers
Kim Kardashian
@kimkardashian
69.5M followers
Virat Kohli
@imvkohli
68.9M followers
YouTube
@youtube
68.6M followers
Bill Gates
@billgates
63.5M followers
The Ellen Show
@theellenshow
62.5M followers
CNN
@cnn
61.9M followers
Neymar Jr
@neymarjr
61.5M followers
X
@x
60.9M followers
Selena Gomez
@selenagomez
60.1M followers