Z.ai: GLM 4.5

GLM-4.5 uses a Mixture-of-Experts (MoE) architecture to dynamically route tokens through specialized expert networks based on input characteristics, enabling efficient processing of 128k-token contexts without proportional latency increases. The MoE design allows selective expert activation per token, reducing computational overhead while maintaining reasoning depth across extended conversations and multi-document analysis tasks typical of agent-based workflows.

GLM-4.5 implements native function calling through a schema-based registry where tools are defined as JSON schemas with parameter constraints, type validation, and description metadata. The model learns to emit structured tool invocations that map directly to function signatures, enabling deterministic tool orchestration without post-processing or regex parsing. Integration with OpenRouter's API exposes this via standard function-calling parameters compatible with OpenAI's format.

GLM-4.5 can be used for batch inference through OpenRouter's API, enabling cost-optimized processing of large numbers of requests. Batch processing typically offers reduced pricing compared to real-time API calls and is suitable for non-urgent inference tasks. The model can process batches of prompts efficiently, with results returned after processing completes. This is valuable for agents running scheduled tasks or processing large datasets.

GLM-4.5 maintains coherent conversation state across turns by encoding prior messages into a compressed representation that persists within the 128k context window. The model uses attention mechanisms to selectively retrieve relevant prior context, enabling agents to reference earlier decisions, tool results, and user preferences without explicit memory management. This is particularly effective for agent workflows where state accumulation (e.g., task progress, discovered facts) must inform subsequent actions.

GLM-4.5 generates code across 40+ programming languages by leveraging training data that includes diverse codebases and syntax patterns. The model understands language-specific idioms, library conventions, and structural patterns (e.g., async/await in JavaScript, type hints in Python, generics in Java) without explicit language-specific modules. Generation is context-aware, respecting indentation, existing code style, and project conventions when completing or extending code snippets.

GLM-4.5 is trained on extensive technical documentation, API references, and code examples, enabling it to understand and reason about complex technical concepts, library APIs, and system architectures. The model can parse API schemas (OpenAPI, GraphQL, Protocol Buffers), understand parameter constraints and type systems, and generate code that correctly uses APIs based on documentation. This is particularly valuable for agent workflows that must interact with external systems.

GLM-4.5 can generate responses that explicitly show reasoning steps, enabling transparency into how conclusions were reached. When prompted with chain-of-thought patterns, the model generates intermediate reasoning steps before final answers, making it suitable for applications requiring explainability or verification. This is implemented through training on reasoning-annotated data and prompt patterns that encourage step-by-step decomposition.

GLM-4.5 supports multiple languages (Chinese, English, and others) with training that enables cross-lingual reasoning — understanding concepts expressed in one language and reasoning about them in another. The model can translate, summarize, and reason across languages without language-specific degradation. This is particularly valuable for global applications and agents that must operate in multilingual environments.

GLM-4.5 supports streaming responses via OpenRouter's API, enabling real-time token generation where tokens are emitted incrementally rather than waiting for full response completion. This is implemented through server-sent events (SSE) or chunked HTTP responses. Streaming is particularly valuable for agent applications where intermediate results must be processed or displayed immediately, and for long-running inferences where latency to first token matters.

GLM-4.5 is trained to follow complex, multi-part instructions with high fidelity, understanding nuanced requirements like output format specifications, tone preferences, and conditional logic. The model maintains instruction adherence even in long contexts where instructions appear early and content appears later. This is implemented through instruction-tuning on diverse prompt patterns and reinforcement learning from human feedback (RLHF) to optimize for instruction-following accuracy.

GLM-4.5 has a defined knowledge cutoff date and can reason about temporal information within its training data. The model understands concepts like 'current year', 'recent events', and 'historical context', enabling it to reason about time-dependent information. However, the model is aware of its limitations and can indicate when information is outside its knowledge cutoff, useful for agents that must handle current events or real-time data.