Matt

no hoax, early preview dari ma-palette-2663f.web.app domain masih WIP, payment masih nyusul, yang mau cobain DM gw with your gmail

@jayair OMAIGAT LET'SGOOOOOOOOOOOOO

DeepSeek coming to Go in 3, 2, 1…

@alibaba_cloud @opencode Go when?

Introducing Qwen3.6-Max-Preview, early preview of our next flagship model ·Improved agentic coding capability over Qwen3.6-Plus ·Stronger world knowledge and instruction following ·Improved real-world agent and knowledge reliability performance Smarter, sharper, still evolving.

@argofowl real flex is no limit at all

codex should remove the weekly usage limit and keep only the 5-hour one that's how you flex your compute advantage

@bridgemindai it needs to hold the "....output less tokens"

OpenAI just priced GPT-5.5 at 2x GPT-5.4. $5 input. $30 output. Claude Opus 4.7: $5 input. $25 output. GPT-5.5 is more expensive than Anthropic's flagship on output tokens. For the first time in two years, OpenAI is the premium-priced option. The model better justify it.

@sama @yacineMTB haven't done much of coding, but in ChatGPT for general use case, it's indeed better

@yacineMTB what about 5.5?

I can't believe how bad gpt 5 was

@jayair yes indeed

DeepSeek V4, what a crazy week man

@opencode OpenCode Go ? :)

Try DeepSeek V4 in OpenCode today 1. `/connect` the DeepSeek provider 2. Grab your key from platform.deepseek.com 3. Select DeepSeek-V4-Pro Tell us what you think

@TencentHunyuan OMGGG don't stop, the era of AI coding will never ends

👋Hi /haɪ/, we're the Tencent Hy /haɪ/ team🐧 Today, we open source Hy3 preview (295B A21B), a leading reasoning and agent model in its size, with great cost efficiency. Give us feedback to help improve Hy3 official! 🤗 hf.co/tencent/Hy3-pr… 📖 hy.tencent.com/hy3-preview

@XiaomiMiMo I don't know why but I'm bullish in mimo, no reason actually, no proof, but just feeling

Introducing MiMo-V2.5 Voice — our full-stack voice lineup for the Agent era. 🚀 Voice AI is moving beyond just hearing and reading — towards precise understanding and flexible expression. That’s why today we’re launching the MiMo-V2.5-TTS Series and open-sourcing MiMo-V2.5-ASR. The TTS series includes three models: 🔹 MiMo-V2.5-TTS — premium built-in voices with fine-grained style control 🔹 MiMo-V2.5-TTS-VoiceDesign — generate brand-new voices from natural-language descriptions 🔹 MiMo-V2.5-TTS-VoiceClone — clone target voices with high fidelity from just a few samples On the recognition side, MiMo-V2.5-ASR is now open source, with strong performance across bilingual speech, Chinese dialects, code-switching, noisy audio, and multi-speaker scenarios. How to get started: For MiMo-V2.5-TTS Series: • It is available for a limited time for free on the Xiaomi MiMo API platform: platform.xiaomimimo.com/docs/usage-gui… • You can also try the TTS models instantly in Xiaomi MiMo Studio: aistudio.xiaomimimo.com/#/c • See more cases at: mimo.xiaomi.com/mimo-v2-5-tts • We’ve also open-sourced MiMo TTS Skills for fast integration into Agent applications: github.com/XiaomiMiMo/MiM… For MiMo-V2.5-ASR: • Demo: mimo.xiaomi.com/mimo-v2-5-asr • Code: github.com/XiaomiMiMo/MiM… • Weights: huggingface.co/XiaomiMiMo/MiM… • Hugging Face Space: huggingface.co/spaces/XiaomiM… Jump in and start building with MiMo-V2.5 Voice today!

@thsottiaux @sandislonjsak Let's gooo

@sandislonjsak I also worked 12 hours to bring it to you. It won't prompt itself (yet), go try it!

I worked 12 hours today and have no energy to try out GPT-5.5. Can anyone tell me is it worth it or not? Thanks!

@sama Imagine a time where ChatGPT is permanently closed, and we focused on Codex, and basically Chat and Repo level coding can be separated in the same plan (just like right now with ChatGPT and Codex)

Also, a ton of new Codex features coming soon! Fun little bundle w/the new model.

@gdb let's try it out

GPT-5.5 is a new class of intelligence. This intelligence makes it intuitive to use; it completes challenging tasks with little micromanagement. Also very token efficient, and runs with low latency and at scale. A real step toward a new way of getting computer work done.

@bridgemindai Every SINGLE DAY, I get a reason to purchase 100$ ChatGPT plan, and a reason NOT to purchase 100$ ChatGPT plan omg

Kimi K2.6 is now #1 on Design Arena. Ahead of Opus 4.7. Ahead of Opus 4.6. Ahead of GLM 5 Turbo. An open-source Chinese model just took the design crown from Anthropic. This is insane.

@songjunkr Let's goooo

2개월 뒤에는 당신의 맥북에서 Opus-4.6 수준의 로컬AI를 구동할수 있을겁니다. 그것의 이름은 Qwen3.7입니다. 이 트윗을 저장하고 정확히 2달뒤에 보세요.

@XiaomiMiMo @ArtificialAnlys Okay now personally I want the unified, all "pro", multimodal model. Thank you.

As Xiaomi’s most powerful model to date, MiMo-V2.5-Pro reaches the level of the world’s leading Agent models across general-purpose reasoning, complex software engineering, and long-horizon tasks as revealed in @ArtificialAnlys Try it now via：platform.xiaomimimo.com/token-plan

@songyoupeng @GoogleDeepMind wow, nice idea actually. Let's see how well it goes

Yay, finally! Introducing Vision Banana🍌 from @GoogleDeepMind, our unified model that outperforms SoTA specialist models on various vision tasks! By treating 2D/3D vision tasks as image generation, we unlock a new foundation for CV. Project page: vision-banana.github.io (1/5)

@bindureddy we're still waiting on deepseek v4 to match this, or the newer glm/qwen/minimax/kimi/etc.

GPT 5.5 costs double of GPT 5.4 More expensive than Opus 4.7 $30 for 1M output tokens 😲

@bridgemindai that much money only to hold the crown for a week, what a time to be alive. but yeah, I'm happy

GPT 5.5 JUST DROPPED. Claude Opus 4.7 is no longer the best model in the world. 82.7% vs 69.4% on Terminal-Bench 2.0. 84.9% vs 80.3% on GDPval. 81.8% vs 73.1% on CyberGym. Not even close. Sam said it would be a leap. He wasn't lying. OpenAI just took the crown back from Anthropic. And they did it across the board. Coding. Security. Math. Agentic tasks. Everything. Claude Opus 4.7 held the throne for exactly 7 days. BridgeBench is running the benchmarks right now.

@thsottiaux let'sgooooo

Introducing GPT-5.5... together with a ton of new Codex features (more on those in the next hour). Included in all paid plans and coming to API soon. Update your Codex app or CLI to use it. openai.com/index/introduc…

OpenCode x Tencent Hy3 preview — free for 2 weeks Strongest in the Hy series - 256K context · reasoning · text only - 295B (A21B) - ☝️means it's built for strong performance at lower token cost; ur welcome