Try For Free
Sign Up
20
GLM-5.3 FlashX is the high-speed variant of Zhipu AI GLM-5.3 Flash, a native multimodal model reaching up to 200 tokens/s on the same hybrid sparse and linear attention architecture, with a 1M-token context window and text, image and video understanding.
GLM 4.5V on AIMLAPI.
GLM 4.6V on AIMLAPI.
GLM 4.7 Flash on AIMLAPI.
GLM 5 Turbo on AIMLAPI.
GLM-5.2 is Zhipu AI flagship MoE LLM with a 1M-token context window, optimized for agentic coding, tool use, reasoning, and long-context tasks.
A latency-optimized GLM-5.2 preview with the same 1M-token context window and 1.5–2× higher output throughput for real-time chat, agents, and coding.
GLM 5V Turbo on AIMLAPI.
GLM-4.5 by Zhipu AI is a high-capacity text-to-text language model featuring a massive 128,000-token context window.
GLM-4.5-Air excels at the intersection of hardware cost-efficiency and high-quality, long-context reasoning capabilities, positioning itself as a highly practical and versatile solution for demanding real-world applications.
GLM-4.6 represents the cutting edge in large language models from Zhipu AI, balancing expansive context capabilities, efficient token use, and strong reasoning performance.
GLM-4.7 is Zhipu AI flagship text LLM optimized for agentic coding and stable multi-step reasoning, supporting long-context workflows.
GLM-5 is Zhipu AI next-generation 745B MoE LLM for high-quality text generation and reasoning, optimized for chat, long-form content, and instruction-following tasks.
Built with a strong emphasis on programming accuracy, structured reasoning, and code reliability, GLM-5-Code delivers consistent performance across code generation, debugging, refactoring, and advanced technical problem-solving.
Z.AI's most powerful language model yet. GLM-5.1 doesn't just generate answers — it plans, executes, iterates, and delivers. Built for long-horizon autonomous tasks with a 200K context window and 128K token output capacity.
GLM-5.3 is Zhipu AI's newest flagship model, offering a 1M-token context window with deep, tunable reasoning and full parallel tool-calling support for building advanced agents.
Zhipu's cost-efficient, natively multimodal Flash model in the GLM-5 family -- 320B total / 18B active parameters, MIT-licensed, 1M-token context. Formerly the stealth "Ox Alpha" model.
Unlike traditional OCR engines that simply output raw text, GLM-OCR preserves formatting, hierarchy, and layout relationships.
