Seven new models launching at Build: let’s go!
Reasoning. Code. Image. Transcribe. Voice.
Built from scratch on a clean data lineage, designed for efficiency, working seamlessly as a family of models
Thread 🧵
#MSBuild
Welcome to Gemini 3.5 Flash, our most powerful model to date. It pushes the frontier of intelligence, speed, and cost putting 3.5 Flash in a class of its own.
We spent the last 6 months making sure Flash is great for real world use cases. It's available everywhere now!
Introducing Antigravity 2.0, a new standalone desktop application that delivers fully on that original glimpse of a truly agent-optimized experience.
Rebuilt from the ground up with multi-agent teams, scheduled tasks, native voice and one-click integration with other Google products.
Learn how to get started with Antigravity 2.0 👇
🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length.
🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models.
🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice.
Try it now at https://t.co/GCdiMzk1Dl via Expert Mode / Instant Mode. API is updated & available today!
📄 Tech Report: https://t.co/drlDrxkYtp
🤗 Open Weights: https://t.co/T13Y8i7SDM
1/n
MiniMax M2.5 just took #2 on the BridgeBench leaderboard.
The scores:
Claude Opus 4.6: 60.1
MiniMax M2.5: 59.7
GPT 5.2 Codex: 58.3
GLM 5: 41.5
MiniMax M2.5 is 0.4 points behind Opus 4.6 and beats GPT 5.2 Codex.
Now look at cost:
Claude Opus 4.6: $5.63
MiniMax M2.5: $0.72
Same performance. 8x cheaper.
100% completion rate.
25.8 second average response time.
This model launched today and it's already outperforming OpenAI's flagship on real world coding tasks.
The Chinese labs aren't just competing anymore.
They're winning on price-to-performance.
MiniMax M2.5 is now BridgeMind certified.
https://t.co/tnVt0Stfj9
Introducing Gemini 3 Flash, our frontier intelligence model, available at scale for everyone. It excels at coding, tool calling, and is stronger than 2.5 Pro across most metrics!! ⚡️
Available in the API at $0.50 in / 1M tokens and $3.00 out / 1M tokens across.
You went 🍌🍌 for Nano Banana. Now, meet Nano Banana Pro.
It’s SOTA for image generation + editing with more advanced world knowledge, text rendering, precision + controls. Built on Gemini 3, it’s really good at complex infographics - much like how engineers see the world:)
Gemini 3 Pro Preview on KingBench (non-agentic):
Gemini 3 Pro scores 100% on KingBench. These are the results from non-agentic tests.
It's ~40% better than Sonnet 4.5 and ~50% better than GPT-5.1 Codex.
I'll be testing it on agentic tests and will report back the scores soon.
We tested Gemini 3 Pro Preview on 5 hard coding/UI tasks against Claude 4.5 Sonnet and GPT‑5.1 Codex.
Scores (our internal rubric):
• Gemini 3 Pro: 72%
• Claude 4.5 Sonnet: 54%
• GPT‑5.1 Codex: 18%
What stood out about Gemini 3 Pro:
• Code feels human: sensible libraries, efficient patterns, minimal prompting.
• Designs are adaptive, not cookie‑cutter
• Consistently correct CDN paths and awareness of newer tools/architectures.
Let's dive deeper.
gemini 3 pro
• our most intelligent model yet
• SOTA reasoning
• 1501 Elo on LMArena
• next-level vibe coding capabilities
• complex multimodal understanding
available now in Google AI Studio and the Gemini API
Introducing Gemini 3 ✨
It’s the best model in the world for multimodal understanding, and our most powerful agentic + vibe coding model yet. Gemini 3 can bring any idea to life, quickly grasping context and intent so you can get what you need with less prompting.
Find Gemini 3 Pro rolling out today in the @Geminiapp and AI Mode in Search. For developers, build with it now in @GoogleAIStudio and Vertex AI.
Excited for you to try it!
An AI powerhouse. Gemini 3 sets a new bar for AI model performance, so you can create, code, and build without missing a beat.
➡️ Multimodal understanding, state-of-the-art reasoning and 1 million-token context window for synthesizing and analyzing videos, images, text, and audio.
➡️ Our best agentic and vibe coding model ever.
➡️ More helpful, better formatted and more concise responses.
Our most anticipated launch of the year is here.
- Gemini 3, our most intelligent model
- Generative interfaces, for perfectly designed responses
- Gemini Agent, made to complete complex tasks on your behalf
See how Gemini 3 can help you learn, build & plan anything 🧵
یہ معاملہ انسانی ہے
غزہ کا محاصرہ توڑ کر انسانی امداد کی راہداری کھلنے تک پے در پے نہتے پر امن شہریوں کے سمندری کاروان غ ز ہ جاتے رہیں گے اور بھوک سے بے حال معصوم جانوں تک امداد پہنچانے کے لیے اپنی جانیں داؤ پر لگاتے رہیں گے
آپ جو گھروں میں ہیں، آپ کا کام ہے ان اساطیل (فلوٹیلاز) کے تحفظ اور ان پر سوار نہتے شہریوں پر دہشتگرد @Israel کی جارحیت کے خلاف آواز اٹھاتے رہیں اور اس سے ہرگز پیچھے نہ ہٹیں۔
#ایڈمن
#GlobalSumudFlotilla
#Hamas
#Gaza
@GovtofPakistan@CMShehbaz