Z.ai

Introducing GLM-5.2: Frontier Intelligence, Open Weights - Significant improvements in coding and agentic tasks - Strong long-horizon capabilities with a 1M context window - Two levels of reasoning effort: GLM-5.2 (max) pushes the limits, while GLM-5.2 (high) strikes a strong balance between performance and token efficiency - MIT-licensed open weights - Same API pricing as GLM-5.1 Tech Blog: z.ai/blog/glm-5.2 Weights: huggingface.co/zai-org/GLM-5.2 API: docs.z.ai/guides/llm/glm… Coding Plan: z.ai/subscribe Chat: chat.z.ai

GLM-5.2 is now selectable in Claude Code via Hugging Face🤗 Inference Providers + hf-claude. Open models are becoming easier to plug directly into real developer workflows. 😀

Follow @zcode_ai for the latest updates. See the full changelog: zcode.z.ai/en/changelog

Introducing ZCode, the official development environment for GLM-5.2 - GLM Coding Plan subscribers: now 1.5x usage quota in ZCode - BYOK supported: works with your existing subscriptions and APIs - Available on macOS, Windows, and Linux Download now: zcode.z.ai/en

GLM-5.2 just tied for #2 on SportEval’s World Cup Exact Score Hit Rate leaderboard. ⚽ 13 exact-score hits out of 76 matches 📊 17.1% accuracy 🥈 Second only to GPT-5.5 That puts @Zai_org ahead of Claude Opus 4.8, DeepSeek V4 Pro, Gemini 3.5 Flash, and more on one of football’s hardest prediction tasks. 🚀 Frontier-level performance, accessible to builders and the open-source community. Follow GLM-5.2’s predictions for the remaining World Cup matches on SportEval: sporteval.ai/arena/world-sc…

Any new features we must have in the next version of glm?

I spend 10 minutes a day trying a new app. matrix.build is my favorite app today. You can tell from a product whether the team's thinking is clear. The thing I love about matrix is it helps you nail down what to build and why before it runs the whole thing. GLM-5.2 now works in it, and the experience is smooth. Great product. Hope it keeps getting more and more polished.

Right alongside Cursor, Devin Desktop (Windsurf) and CLI now support GLM-5.2 as well. FrontierCode Extended is a benchmark we care deeply about for real-world engineering tasks, so it's great to see GLM-5.2 coming close to GPT-5.5 here. Also congrats to @Kimi_Moonshot!

GLM-5.2 is now available in Cursor. The model has performed strongly on OpenRouter's Cursor usage rankings over the past week. Would love to hear comparisons of the experience using BYOK (GLM Coding Plan, etc.), Cursor's built-in option, and calling it through OpenRouter.

GLM-5.2 is available in Perplexity's Agent API. Just tested it, and it's powerful when paired with the Search SDK inside a sandbox. - Spin up a sandbox environment - Call the web search tool (people search, finance search, etc. also work) - Pretty good for wide & deep research

Here is the prompt method behind this AR try-on app. The trick is not a magic prompt. It is the architecture of the prompt, and it works across GLM-5.2 and other frontier models. Full prompt: chat.z.ai/space/k148m5py… The prompt has two parts. First, a task description. You write this fresh for each app to define the business logic. Second, a five-round polish process: Round 1 through Round 5. The structure is fixed and reusable across any app, but the specific content of each Round is tailored to the app at hand. The flow is simple. The task description builds the functional skeleton, then the five Rounds run in sequence to refine it into something that looks like a finished product. Why split it this way? Single-pass generation always prioritizes "it runs" over "it looks good." So it's better not to chase one perfect prompt. Divide the work. The business description makes it function. The five Rounds make it look like a real product. The polish is a reusable pipeline, not something you reinvent every time, even though you fill in app-specific details each time. How to write the task description: Treat it as a real PRD and engineering spec, not a user wishlist. Include the tech stack, information architecture, module specs, API integration, data model, and acceptance criteria. Declare autonomy at the top. State that the model should not ask questions, not stop early, and verify its own work. Otherwise it will pause to ask and break the long task. Write the fallback paths explicitly. Cover unsupported devices, older OS versions, and offline states. If you skip this, the model improvises at the edges and crashes. Number your acceptance criteria. Each should be independently verifiable, for example "tap a product and the look changes within 0.5 seconds." The principles behind the five Rounds: Quantify "good" into numbers. Models execute poorly on adjectives and precisely on constraints. Use spring response 0.3 to 0.4, button scale 0.95 to 1.0, at most 5 font sizes, and sound effects under 200ms. These principles stay constant, even as the exact targets shift per app, which is why the structure can stay fixed. List what is forbidden. Models cut corners in predictable ways, such as gray placeholders, solid color blocks, and spinners. Name them directly with "DO NOT" and provide an acceptable fallback. Inventory before fixing. Each Round follows the same loop: audit every asset, verify it is not a placeholder, replace, amplify, and re-screenshot to confirm. Strip the "tutorial" feel. AI output gives itself away with faker text, .test links, and emoji-only empty states. The final Round removes these.

GLM 5.2 x AWS GLM-5.2 is accessible via Z.ai GLM API on AWS Marketplace 🚀 Powerful long-horizon autonomous workflows, top-tier coding & multi-step agent reasoning capability, delivered through a single unified API endpoint. Integrate seamlessly within your AWS cloud without self-host GPU maintenance. aws.amazon.com/marketplace/pp… #GLM52 #ZhipuAI #AWSMarketplace #EnterpriseLLM

@didier_lopes Thanks for supporting @slime_framework ❤️

Incredible how Z. ai literally has their RL infrastructure open source. The entire OPD post-training of GLM-5.2 took on this slime platform took ~2 days. github.com/THUDM/slime

"...at least as good as Opus 4.8 and GPT 5.5."

GLM-5.2 is not only stronger on benchmarks, but also much better in real app development scenarios — iOS, Android, WeChat Mini Programs, and more. Behind this jump is a full loop from environment construction, evaluation, data optimization, reward design, to training. Real tasks, real execution, real improvement.

Long-horizon is more than a concept. It should live in real-world scenarios, empowering AI builders to solve the problems that matter. And more scenarios are on the way.

GLM-5.2 delivers a substantial leap in app development capabilities, which also represent demanding long-horizon tasks. Results: - GLM-5.1: 21/70 - GLM-5.2: 48/70 - Claude Fable 5: 56/70 That's more than a twofold improvement from GLM-5.1 to GLM-5.2. These come from an internal benchmark of 35 challenging mobile development tasks, each run twice for a total of 70 trials. We measured task completion, defined as core features working without major issues.

Announcing AA-Briefcase, the benchmark for the next era of agentic knowledge work AA-Briefcase is our new benchmark for testing models on long-horizon knowledge work tasks in complex projects built by industry experts. Models are evaluated on multi-week projects, each with many linked tasks and thousands of input source files. We evaluated Claude Fable 5 from @AnthropicAI before it became unavailable, and it currently leads with an Elo score of 1587, followed by Claude Opus 4.8 (max, 1356), Opus 4.7, and the recently-released GLM 5.2 (max, 1266) from @Zai_org. Claude Fable 5 cost $31 on average to run each AA-Briefcase task, followed by Claude Opus 4.8 at $10.40, GPT-5.5 (xhigh) at $3.68 and GLM-5.2 (max) at $2.40. AA-Briefcase comprises four private scenarios, each representing a multi-week knowledge work project set in a realistic organizational context. A public fifth scenario has been released via @huggingface as a representation of scenario structure, submission, and grading (AA-Briefcase Lite). This does not count toward official AA-Briefcase results, and is demonstrative only. Key elements of AA-Briefcase: ➤ Realistic long-horizon projects: AA-Briefcase moves beyond single, disconnected prompts by evaluating models across a coherent long-horizon project. Tasks build week by week, draw on shared institutional context, and require deliverables such as financial models, board presentations, and design mock-ups ➤ Large volumes of fragmented context: AA-Briefcase requires models to reason across thousands of inputs, including company documents, meeting transcripts, large-scale data exports, 25,000+ Slack messages and 3,500+ emails. These sources are fragmented, messy, and often contain realistic contradiction, testing whether models can navigate the ambiguity of real-world knowledge work ➤ Composite rubric and pairwise grading: AA-Briefcase combines binary rubric checks for ground-truth correctness with pairwise grading on analytical quality and presentation quality. Unlike many evaluations that focus on a single metric, AA-Briefcase tests agentic capabilities more comprehensively, exposing cases where models produce outputs that look polished but are incorrect or lack analytical rigor ➤ Built by industry experts: AA-Briefcase scenarios mirror real-world knowledge work, with tasks developed over months by experts across data science, product management and corporate strategy from companies including Google, McKinsey & Company and BCG. Task challenges are drawn from professional experience, making AA-Briefcase more reflective of the ambiguity, messy context and competing priorities that define real-world knowledge work Key results: ➤ Claude Fable 5 leads AA-Briefcase at 1587 Elo: This is followed by Claude Opus 4.8 (1356) with the next-best non-Anthropic model, GLM-5.2 (max), ~90 points back at 1266. Note that Claude Fable 5 did not use the Opus 4.8 fallback for any task in AA-Briefcase ➤ Cost per task varies by ~800x across models tested: Claude Fable 5 leads the benchmark but costs more than $31 per task on average, compared to ~$0.04 for DeepSeek V4 Flash (max). The strongest price/performance options are open weights models such as GLM-5.2 (max) and DeepSeek V4 Pro (max), with GLM-5.2 (max) scoring only ~90 Elo below Claude Opus 4.8 (max) for less than 25% of the cost ➤ Real-world complexity remains difficult for models: The top performer, Claude Fable 5, satisfies all rubric criteria on just 3% of AA-Briefcase tasks. On 31 of 91 tasks, no model scores above 50% on the rubric criteria ➤ Task difficulty scales with the number of required input files: For each rubric check, we identify the set of source files needed to pass. Across all models, pass rates fall as this file count increases, though top-tier models degrade less than weaker models More details below in thread ⬇️

GLM-5.2 can now run locally in llama.cpp and Unsloth Studio. Check the graph below for the accuracy of each GLM-5.2-GGUF quantization. Full guide: unsloth.ai/docs/models/gl…

GLM-5.2 is free when used with Hugging Face Inference Providers for the next 5 hours: huggingface.co/zai-org/GLM-5.…

GLM-5.2 can now be run locally!🔥 The 2-bit model retains ~82% accuracy after we shrunk it from 1.51TB to 238GB (-84% size). Run on a 256GB Mac or RAM/VRAM setups. GLM-5.2 is the strongest open model to date. Guide: unsloth.ai/docs/models/gl… GGUF: huggingface.co/unsloth/GLM-5.…