Pro/zai-org/GLM-4.7

GLM-4.7 is the flagship text model launched by Zhipu, with a core focus on "strengthening agentic coding capabilities to cover complex task solving and efficient creation across multiple scenarios." It can autonomously plan, execute, and verify code tasks.

• Key Upgrades: Compared to GLM-4.6, its mathematical and reasoning abilities have improved by 41%, achieving a HLE benchmark score of 42.8%, surpassing GPT-5.1. It introduces reserved thinking and round-based thinking modes, ensuring more stable execution of complex tasks and allowing users to control inference costs as needed.
• Applicable Scenarios: Covers multiple professional scenarios, including agentic coding, multimodal interaction app development, frontend design, office creation, deep research, software development, algorithm implementation, and script automation.
• Evaluation Data: The τ²-Bench interactive tool invocation evaluation scored 84.7 (open-source SOTA), and the BrowseComp web task evaluation scored 67.
• Product Value: Generates complete, runnable code frameworks, reducing manual assembly and debugging costs, and accelerating the process from idea to implementation.
• Core Parameters: Context window size: 200K; maximum output tokens: 128K; supports structured format outputs such as JSON.
• Evaluation Data: Supports over 100 programming languages, with an average of 3.2 toolchain calls per task, and an autonomous repair rate exceeding 65%.

───────────────────────────────────────────────────────────────────

Core Capabilities

💻 Strong breakthrough in agentic coding: From requirement understanding to solution breakdown, it autonomously completes multi-tech-stack integration, generating structurally complete, runnable code that covers front-to-back end collaboration scenarios.

🧠 Controllable thinking modes: Supports interleaved, reserved, and round-based thinking. For simple tasks, it reduces thinking latency; for complex tasks, it increases accuracy by enabling thinking; and for long-term tasks, it provides more efficient caching.

🎨 Enhanced frontend aesthetic quality: Accurately understands UI design specifications, providing high-quality default solutions for layout, color schemes, and component styles, thus reducing the time spent on fine-tuning styles.

📝 Immersive creation experience: Text expressions are delicate and vivid, character roles adhere consistently to their established personas, plot progression feels natural, and office creation aligns with mainstream formatting standards.

───────────────────────────────────────────────────────────────────

Test Data