Alex

Plugin-based architecture @openclaw is a massive W. No more bloated binaries! 📦 Just installed the Codex harness as a standalone plugin - clean, fast & modular. This is how pro CLI tools should evolve. Kudos to @steipete & team for the heavy lifting! 🛠️ #OpenClaw #AI #DevTools

@sama Stop getting better at sounding right and start getting better at knowing when you're wrong

what would you most like to see improve in our next model?

@Sedrak1010 @openclaw @steipete Eagle eye! 🦅 Spot on. It’s a classic trade-off: deep context vs. snappiness. I keep it off for routine tasks to avoid the lag, only toggling it for complex dives. Hopefully, future RAG optimizations will make it feel as lightweight as the core! 🛠️ @openclaw

@oleksiyML @openclaw @steipete Noticed your active memory is off too. It's great and I wish it wasn't making the system sluggish.

Plugin-based architecture @openclaw is a massive W. No more bloated binaries! 📦 Just installed the Codex harness as a standalone plugin - clean, fast & modular. This is how pro CLI tools should evolve. Kudos to @steipete & team for the heavy lifting! 🛠️ #OpenClaw #AI #DevTools

@YAHHPP @steipete @openclaw Basically, @openclaw went on a diet Instead of one giant 500MB+ package with binaries for every platform, we now have a 'lean core'. You only 'plug in' the specific harnesses or providers you actually use. It’s faster, lighter, and keeps your environment clean. Modular is the way

@oleksiyML @steipete @openclaw Someone needs to explain more! Lol

@jmanbuilds @steipete @openclaw Codex is the solid workhorse. Claude Code is great, but currently feels like an extreme sport😅Anthropic is playing a 'Seek&Destroy' game with anything using their name lately Using it is like wearing a 'Ban Me' sign to dinner. Stick with Codex if you like your access...stablе😉

@oleksiyML @steipete @openclaw I'm looking at setting up a Claude code harness for my agent. Which is better? Codex or Claude code?

@this_subhan @steipete @openclaw Exactly. It turns the CLI from a Swiss Army knife that's too heavy to carry into a precision toolset you actually want to use every day. Scalability is the name of the game in the 2026 AI era.

@oleksiyML @steipete @openclaw Plugin-based architecture is definitely the right direction 👏 modular, lightweight, and way more scalable. Big win for dev tooling.

@Layton_Gott @steipete @openclaw Truly! It feels like we’re shipping in dog years. 🐕 This modular shift is exactly what was needed to keep that pace without the whole thing collapsing under its own weight. The velocity is insane.

@oleksiyML @steipete @openclaw OpenClaw is moving so freaking fast…

@KorduGG @openclaw @steipete It’s all about a lean core now. You get the base CLI, and then just 'plug in' what you actually need-harnesses, providers, specific tools. It keeps the binary small and the startup time lightning fast. Check > openclaw plugins --help to see the magic! ✨

@oleksiyML @openclaw @steipete Interesting how does this work?

Gemini API Webhooks are a game changer for agentic workflows. Polling was the 'dark ages' of DevX. Now we can finally build elegant, long-running async tasks without the overhead. 🏗️ Native support for GenMedia and batch jobs feels right. Great job @OfficialLoganK! 👏

@downandoutlaw @mweinbach $10 per user per month with Google AI Pro $100 per user per month with Google AI Ultra

@mweinbach I have AI pro. Do I have to enable this every month to get the credit or just once?

Heads up for OpenClaw and Google users, Google gives you a $10 GCP credit per month for AI Pro and $100 per month for AI Ultra to use on the Gemini API Don't use sketchy oauth hacks to pull inference from other pools, just enable this developers.google.com/program/my-ben…

@openrouter Smells like a Chinese model under the hood.

🥷 New stealth model: Aurora Alpha Aurora Alpha is an extremely fast reasoning model. It is built for coding assistants, with fast completions to keep you in the flow, and is extremely powerful for real time conversational applications.

@AntLingAGI Is it possible to use Ling Studio without having a Chinese phone number?

🚀 Ling Studio is officially live — the official web platform to interact with Ant Ling models, offering 500,000 free tokens every day to explore, test, and build. In Ling Studio, you’ll experience Ling-1T’s lightning-fast responses, explore Ring-1T for powerful and stable reasoning, and try Ming-flash-omni-Preview to explore image and audio understanding - all in one place. Key features: * Flexible Ling / Ring configuration (parameters + system prompts) * Native tool calling, including web search * API-ready for easy integration into your applications Coming soon: file-based chat, image generation, a growing set of model skills, and more.

@Ali_TongyiLab (•_•) (■_■) <) )🤜 🤛( (> / \ / \

(•_•) < New Year gift <) > is on the way! / \

@XxMakio OH NO... IT WORKED?

Hey guys, can you use edit image to test if you can remove the watermarks here? (It's impossible)

My 2025 OpenRouter Wrapped is here! 🏆 Top 6% of Gemini 2.0 Flash Thinking Experimental 01-21 users. #OpenRouterWrapped openrouter.ai/wrapped/2025/7…

@Meituan_LongCat Video Continuation modes & Infinite-length: This is what everyone is missing. Everyone can generate beautiful 5 seconds. No one can generate 5 minutes without the hero's face turning into mush.

Meet LongCat-Video-Avatar: a robust audio-driven avatar model that pushes the boundaries of long-form video generation. Compared with the previous InfiniteTalk, LongCat-Video-Avatar delivers far better long-sequence stability and realism. New highlights： ⚙ Built on the LongCat-Video architecture, now supporting Audio-Text-to-Video (AT2V), Audio-Text-Image-to-Video (ATI2V), and Video Continuation modes. 🎭 Open-source SOTA Realism: Ranked 1st in overall anthropomorphism scores for both single and multi-subject scenarios in EvalTalker evaluations (492 participants, 3 independent raters per video). ♾ High-quality long videos: Cross-Chunk Latent Stitching prevents pixel degradation and error accumulation over time, ensuring seamless stitching quality. 🔒 Long-term consistency: Reference Skip Attention maintains ID consistency while eliminating rigid copy-paste artifacts. 🪄 Supports multi-person and infinite-length video generation. 🔗Open-sourced Code: github.com/meituan-longca… Hugging Face: huggingface.co/meituan-longca… Project: meigen-ai.github.io/LongCat-Video-… Paper: github.com/meituan-longca…

@Meituan_LongCat made by @Meituan_LongCat for @NanoBanana in the longcat.chat

🚀 LongCat-Image is now open-source! A bilingual (CN/EN) image generation & editing model with exceptional consistency, clarity, and text precision. 🔄 Superior editing performance with accurate visual alignment 📝 Renders complex Chinese characters, incl. rare glyphs 🖼️ Realistic rendering powered by data strategy and RL training ⚡ Fast, low-VRAM, lightweight inference ⚙️ 6B hybrid DiT architecture achieves 20B+ MoE performance 📊 Objective benchmarks prove it! LongCat-Image leads open-source: - Image Editing: ImgEdit 4.50 (open-source SOTA), GEdit CN/EN 7.60/7.64 rivals commercial ones; - Text Rendering: ChineseWord 90.7 outclasses all competitors; - Text-to-Image: GenEval 0.87 & DPG 86.8 is competitive with top-tier open-source & closed-source models. We release multiple LongCat-Image versions (mid/post-training model) with the full training toolchain to support multilingual visual research. 📄 GitHub: github.com/meituan-longca… 🤗 HF: huggingface.co/meituan-longca… 📑 Report: github.com/meituan-longca…

@Meituan_LongCat Mind-blowing models! 🚀 Built an open-source bridge so Apple devs can plug them straight into Xcode 26 - keep up the epic work! Repo in profile if anyone needs it. UPD: 500k tok/day free, or just grab 5M with one form (yep, I'm already approved!)

🔥 LongCat API Platform is LIVE! 🆓 Developers: Claim your 100,000 FREE tokens per day and leverage our cutting-edge LLM capabilities. 🛠️ LongCat API Platform is compatible with two leading API formats - choose OpenAI API or Anthropic API based on your needs! Build, test, and deploy AI solutions with unmatched efficiency. 🚀 Start building today: longcat.chat/platform

@Meituan_LongCat Mind-blowing models! 🚀 Built an open-source bridge so Apple devs can plug them straight into Xcode 26—keep up the epic work! Repo in profile if anyone needs it.

🚀 LongCat-Flash-Thinking: Smarter reasoning, leaner costs! 🏆 Performance: SOTA open-source models
on Logic/Math/Coding/Agent tasks 📊 Efficiency: 64.5% fewer tokens to hit top-tier accuracy on AIME25 with native tool use, agent-friendly ⚙️ Infrastructure: Async RL achieves a 3x speedup over Sync frameworks 🔗Model: huggingface.co/meituan-longca… 💻 Try Now: longcat.ai

@Meituan_LongCat Mind-blowing models! 🚀 Built an open-source bridge so Apple devs can plug them straight into Xcode 26 - keep up the epic work! Repo in profile if anyone needs it.

🎉LongCat's Latest Updates: 1️⃣ API Access: LongCat-Flash-Thinking is now available via API (Claude Code integration compatible) 2️⃣ Daily Free Quota: Increased from 100,000 → 500,000 tokens/day 3️⃣ Apply for More: Submit request on "Usage Info" page for 5M tokens/day after approval 4️⃣ New Docs: Change Log & Claude Code Configuration 🔄 Updated: API Docs, Quick Start Guide, FAQ Explore the speed & power of LongCat with ease ⚡ 🔗 LongCat API Platform: longcat.chat/platform/usage 🤗 HuggingFace: huggingface.co/meituan-longca… 💻 GitHub: github.com/meituan-longca…