Z.ai

Introducing GLM-5-Turbo: A high-speed variant of GLM-5, excellent in agent-driven environments such as OpenClaw. Coding Plan Max: z.ai/subscribe OpenRouter: openrouter.ai/z-ai/glm-5-tur… API: docs.z.ai/guides/llm/glm…

Don't panic. GLM-5.1 will be open source.

Introducing GLM 5 Turbo for understanding research papers 🚀 Highlight any section of a paper to ask questions and “@” other papers for quick context, comparisons, and benchmark references

GLM-5 Turbo, a new fast model based on @Zai_org's GLM-5 is live on OpenRouter! In the last few days, GLM-5 Turbo has picked up significant usage in @OpenClaw. Use it now to speed up your agents!

glad to have a deep chat with @steipete about openclaw about its impact, mission, what it means for people, and how @Zai_org can contribute together. Let’s continually.. 🦞Build for open source. 🦞Do meaningful work. 🦞Take OpenClaw everywhere .. to every nation, every school, every human.

❤️ the sunshine, the sense that everyone is here because they genuinely care about building the future a group of purists, and a group of people the world may still be underestimating

My vibe check on the AI model race. It’s a hierarchy of ambition, where every lab strives for the highest level they can reach: Level 1: The Disruptors. They don't just build models; they rewrite the industry's playbook. They define the paradigm shifts that move the entire human race forward. They are the roadmap. Level 2: The Powerhouses. The elite performers. They consistently deliver best-in-class capabilities and top-tier benchmarks. If you want reliable "big brains" available, you go to them. Level 3: The Specialists. They carve out a unique craft (high speed, a distinct personality, or a specialized vertical edge). They aim to be someone’s indispensable favorite. Level 4: The Philanthropists. When the top three levels are out of reach, they lead with the heart. By empowering the community and opening doors for others, generosity becomes their path to relevance. The goal is always to ascend. If you can’t reinvent the world, aim for the top tier. If you can’t hit the benchmarks, perfect your craft. If all else fails, be the one who gives.

🤗To celebrate @NVIDIAGTC and the launch of GLM-5-Turbo, we're running a giveaway! Our team will be picking lucky winners on a rolling basis. You have a chance to get a free month of the Max Coding Plan! Here's how to join in: a retweet, reply, or post with: 1. GLM-5-Turbo usecases or 2. A short write-up based on your experience or honest take The window closes in 48 hours. All prizes will be sent out within 72 hours after that. Feel free to get your User ID here: z.ai/subscribe Please jump in, we'd love to see what you're building! ps. @louszbd is at GTC right now.😆 Come say hi in person!

Support an interesting multi-AI social ecosystem. Every rule is designed by Claw. github.com/agi-bar/clawco…

Kilo Coders are head over heals for GLM-5-Turbo! Fresh out of the gate, it is already leading the charge in KiloClaw 🦞🏇🦞🏇🦞

GLM-5-Turbo just released Available in Crush now No update required

Usage limits tripled for GLM-5-Turbo in GLM Coding Plan! Enjoy the same high-volume capacity as GLM-4.7 during non-peak hours. Availability: Anytime except 2–6 AM ET. Ends: April 30.

pony-alpha-2 has finally leveled up into GLM-5-Turbo can’t wait to see how it performs! DM me with your User ID if you need a rate limit increase for GLM-5-Turbo docs.z.ai/guides/llm/glm…

New experiment

Note: As an experimental version, GLM-5-Turbo is currently closed-source. All capabilities and findings will be incorporated into our next open-source model release.

Need it sooner? Apply for Early Access: - Pro (GLM-5-Turbo): docs.google.com/forms/d/e/1FAI… - Lite (GLM-5): docs.google.com/forms/d/e/1FAI…

Introducing GLM-5-Turbo: A high-speed variant of GLM-5, excellent in agent-driven environments such as OpenClaw. Coding Plan Max: z.ai/subscribe OpenRouter: openrouter.ai/z-ai/glm-5-tur… API: docs.z.ai/guides/llm/glm…

GLM-OCR has accumulated over 3M downloads. We are releasing its technical report: arxiv.org/abs/2603.10910 We welcome your feedback!

🛠️ Builders & devs — the Build with TRAE & Z.ai Creative Lab series is live! @Trae_ai Hands-on AI hacking in Fukuoka, Boston, Chicago, San Diego . Live demos, guided prototyping with TRAE + Z.ai tools, lightning pitches, and real awards. 🌟Sign up here ： • Fukuoka (Mar 19) → luma.com/vndjyrmv • Boston (Mar 22) → luma.com/y5fnd2g3 • Chicago (Mar 24) → luma.com/8ovatja7 • San Diego (Mar 31) → luma.com/c803h31l

🧵 1/4 Still waiting for DeepSeek-V4? We (@Zai_org) made DSA 1.8× faster with minimal code change — and it's ready to deliver real inference gains on GLM-5. IndexCache removes 50% of indexer computations in DeepSeek Sparse Attention with virtually zero quality loss. On GLM-5 (744B), we get ~1.2× E2E speedup while matching the original across both long-context and reasoning tasks. On our experimental-sized 30B model, removing 75% of indexers gives 1.82× prefill and 1.48× decode speedup at 200K context. How? 🧵👇 #DeepSeek #GLM5 #Deepseekv4 #LLM #Inference #Efficiency #LongContext #MLSys #SparseAttention