Z.ai

GLM-5.3-Flash

Z.ai

GLM-5.3-Flash is a natively multimodal GLM model built on a newly trained base. Its hybrid sparse and linear attention, manifold-constrained hyper-connections, and 30T-token multimodal pretraining target strong capability at lower long-context serving cost. Native visual perception supports agents that observe interfaces, inspect rendering results, and refine code. The 320B model activates 18B parameters per token and supports million-token contexts. MIT weights and official API/chat access provide both local and hosted options.

GLM-5.3

Z.ai

GLM-5.3 advances coding, long-horizon engineering, and vulnerability-analysis capabilities through post-training on the GLM-5.2 base model. The model remains text-based and supports a million-token context, selectable reasoning effort, and persistent thinking history. Z.ai initially announced hosted availability before releasing downloadable weights. Those weights use the custom GLM-5.3 License rather than the MIT license of earlier GLM-5 models. It is distinct from GLM-5.3-Flash, which introduces a new native multimodal architecture.

GLM-5.2

Z.ai

GLM-5.2 expands the GLM engineering line to a million-token context and improves reliability on sustained coding tasks. IndexShare reuses sparse-attention indexing across groups of layers to lower long-context computation, while multi-token prediction improvements accelerate decoding. Training emphasizes maintaining goals and using noisy repository and tool context over extended sessions. Public MIT weights and official API access support deployment. The million-token limit is an input-context capability, not a guarantee of reliable retrieval or correct execution at every length. GLM-5.2 Fast is a faster-serving route with the same weights, grouped here rather than cataloged as a separate trained model.

GLM-5V-Turbo

Z.ai

GLM-5V-Turbo is Z.ai’s hosted multimodal coding foundation model. It processes text, images, videos, and rendered file content for visual coding, long-horizon planning, and tool-based action execution. Typical workflows include recreating interfaces from screenshots, inspecting rendered pages while debugging, and exploring websites through an agent. Official documentation lists a 200K context and 128K maximum output. A technical report describes the native multimodal model; no public downloadable checkpoint is established for this specific Turbo tier.

GLM-5.1

Z.ai

GLM-5.1 strengthens the GLM engineering model’s ability to stay productive over extended agent sessions. Post-training improves problem decomposition, experiment execution, tool use, and strategy revision across hundreds of rounds. It targets repository generation, complex debugging, and long-horizon software optimization. MIT weights and an official API permit hosted and local use. This page dates the verified public model release to the official April 7 announcement; earlier subscription rollouts are not established by the cited release record. The March 27 official Coding Plan announcement establishes public subscription access before the April 7 broader release.

GLM-5-Turbo

Z.ai

GLM-5-Turbo is a hosted language model optimized during training for OpenClaw agent workflows. It strengthens tool invocation, instruction decomposition, scheduled and persistent tasks, and long-chain execution. A roughly 200K context and up to 128K output support multi-step agents, with function calling, JSON structured outputs, context caching, and MCP integration. Official documentation provides a named API endpoint. The weights and detailed architecture are not publicly established; it should not inherit open-weight status from the standard GLM-5.

GLM-5

Z.ai

GLM-5 scales the GLM family to 744B parameters and introduces DeepSeek Sparse Attention for efficient long-context inference. It targets complex systems engineering and long-horizon agents, with expanded pretraining and asynchronous reinforcement learning using the slime infrastructure. The model accepts and generates text, including structured tool calls. MIT weights support self-hosting, and official API/chat products provide hosted access. The February 11 public weight release precedes the February 12 dated product blog and documentation announcement.

GLM-4.7-Flash

Z.ai

GLM-4.7-Flash is the lightweight, free-tier member of the GLM-4.7 language-model line. It targets coding, reasoning, writing, translation, and high-frequency conversational workloads at lower inference cost. The 30B MoE checkpoint activates roughly 3B parameters per token and supports a roughly 200K context. Public MIT weights enable local deployment, while Z.ai exposes hosted Flash access. The FlashX hosted service variant is represented here rather than as a duplicate model page.

GLM-4.7

Z.ai

Advanced coding-focused language model with significant improvements in multilingual agentic coding, terminal-based tasks, UI generation, tool using, and complex reasoning. Features Interleaved Thinking, Preserved Thinking, and Turn-level Thinking capabilities for enhanced stability in complex multi-turn tasks. Achieves substantial gains over GLM-4.6 across coding benchmarks, web browsing, and mathematical reasoning.

AutoGLM-Phone

Z.ai

AutoGLM-Phone is a released vision-language model specialized for mobile-device operation. It interprets phone screenshots and natural-language task instructions to plan actions such as taps, scrolling, and app navigation. The open-source Phone Agent system supplies ADB execution around the model; the 9B model weights are a distinct artifact and are represented here. Chinese and multilingual English/Chinese checkpoints are grouped as variants. MIT model weights and code enable local mobile-agent experiments, with deployment requiring a compatible device and execution framework.

GLM-4.6V-Flash

Z.ai

GLM-4.6V-Flash is the compact tier of the GLM-4.6V native multimodal tool-use family. It targets lower-latency or resource-constrained deployments while retaining image/video/document understanding and a 128K context. Visual inputs can participate directly in agent tool workflows, including screenshot-based interface reconstruction and document extraction. MIT weights permit self-hosted use; the official provider also documents hosted access. Capabilities and benchmark results should be attributed to this compact checkpoint rather than copied from the larger flagship.

GLM-4.6V

Z.ai

GLM-4.6V is the flagship vision-language tier introducing native multimodal tool use in the GLM-V line. Images and other visual context can pass through tool calls without first being reduced to text descriptions. The 128K context supports long documents and videos, while visual reasoning enables screenshot-to-code reconstruction and mixed-media document workflows. Public MIT weights and official hosted access are available. The 9B Flash tier is separately cataloged because it has a materially different deployment footprint.

GLM-4.6

Z.ai

GLM-4.6 is a substantive update to the GLM agentic language-model line, improving software engineering, tool use, and reasoning. Its context expands from the previous 128K class to approximately 200K tokens, allowing larger code and document workflows. The release emphasizes reliable multi-step agent execution and practical developer integrations. MIT weights and official API/chat access make it available for self-hosting and hosted use. It is a text model; the independently released vision-language GLM-4.6V family is cataloged separately.

GLM-4.5V

Z.ai

GLM-4.5V builds visual reasoning and agent capabilities on GLM-4.5-Air-Base. A vision encoder and MoE language backbone process images, videos, charts, documents, and GUI screenshots. The model supports a switch between direct response and thinking behavior for latency-sensitive or more difficult visual tasks. MIT weights, official API access, and demonstration applications support local and hosted use. Its desktop assistant is a separate execution harness around the model.

GLM-4.5-Air

Z.ai

GLM-4.5-Air is the smaller, separately named tier of the GLM-4.5 release. It combines reasoning, code generation, and native function calling while activating 12B parameters out of 106B total. Its 128K context and thinking-mode switch support coding and agent deployments with a lower inference footprint than the flagship. MIT base and post-trained weights are available, alongside official hosted API access. Sparse activation lowers per-token compute but does not remove the need to store the complete weights.

GLM-4.5

Z.ai

GLM-4.5 unifies reasoning, coding, and agentic tasks in one foundation model. Thinking and non-thinking modes offer a tradeoff between deeper reasoning and direct responses. Native function calling supports external tools, and the 128K context handles documents and software repositories. Public MIT weights include both base and post-trained checkpoints, with hosted access through Z.ai services. The lower-compute Air tier has a separate page; ordinary precision variants and the hosted Flash alias are grouped here.

GLM-4.1V-Thinking

Z.ai

GLM-4.1V-Thinking is the reasoning-focused GLM vision-language model built from GLM-4-9B-0414. Curriculum-based reinforcement learning improves visual reasoning, image and video questions, document understanding, and GUI tasks. It supports arbitrary image aspect ratios and images up to 4K resolution within a 64K context. The public 9B Thinking and Base checkpoints are grouped as variants under MIT terms. The model is a substantive multimodal reasoning release despite its comparatively compact language backbone.