GLM-4 Models by Zhipu AI
Last refreshed 2026-07-11. Next refresh: weekly.
Details
Capabilities
About
The GLM-4 family, developed by Zhipu AI and Tsinghua University, represents an evolving series of large language models renowned for their multilingual capabilities and state-of-the-art performance. Building upon previous ChatGLM generations, these models are pre-trained on an extensive dataset of ten trillion tokens across Chinese, English, and 24 other languages. They undergo rigorous multi-stage post-training, involving supervised fine-tuning and reinforcement learning from human feedback, which enables them to rival or surpass GPT-4 on various benchmarks. The series includes versions like GLM-4, GLM-4-Air, and GLM-4-9B, each tailored for different tasks and resource constraints. A notable feature is the GLM-4 All Tools model that can autonomously use web browsers and Python interpreters for complex task completion. Open-source variants, such as GLM-4-9B and its chat-optimized version, along with multimodal models like GLM-4V-9B, which integrates image processing, highlight the family's versatility.
Decision facts
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs vision, reasoning, and multimodal inputs.
Use when the workload needs vision, reasoning, and multimodal inputs.
Use when the workload needs 200K context, JSON / Tool use, and structured outputs.
Use when the workload needs 128K context, JSON / Tool use, and multimodal inputs.
Use when the workload needs 64K context, JSON / Tool use, and multimodal inputs.
Use when the workload needs 128K context and 9B parameters.
Use when the workload needs 128K context, 32B parameters, and structured outputs.
Use when the workload needs 128K context and structured outputs.
Use when the workload needs 128K context and structured outputs.
Use when the workload needs 198K context and structured outputs.
Use when the workload needs 198K context, reasoning, and JSON / Tool use.
Use when the workload needs 128K context and 9B parameters.
Use when the workload needs JSON / Tool use, structured outputs, and prompt caching.
Use when the workload needs realtime voice, multimodal inputs, and audio.
Use when the workload needs 128K context, 9B parameters, and multimodal inputs.
Use when the workload needs vision and multimodal inputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| GLM-4.1V-Thinking-Flash | Use when the workload needs vision, reasoning, and multimodal inputs. | Unknown release | visionreasoningmultimodal inputs | Current |
| GLM-4.1V-Thinking-FlashX | Use when the workload needs vision, reasoning, and multimodal inputs. | Unknown release | visionreasoningmultimodal inputs | Current |
| GLM 4.7 | Use when the workload needs 200K context, JSON / Tool use, and structured outputs. | 2026-03 | 200K contextJSON / Tool usestructured outputs | Current |
| GLM 4.6V | Use when the workload needs 128K context, JSON / Tool use, and multimodal inputs. | 2026-02 | 128K contextJSON / Tool usemultimodal inputs | Current |
| GLM 4.5V | Use when the workload needs 64K context, JSON / Tool use, and multimodal inputs. | 2026-01 | 64K contextJSON / Tool usemultimodal inputs | Current |
| GLM-4 Code 9B | Use when the workload needs 128K context and 9B parameters. | 2025-05 | 128K context9B parameters | Current |
| GLM-4 32B | Use when the workload needs 128K context, 32B parameters, and structured outputs. | 2025-03 | 128K context32B parametersstructured outputs | Current |
| GLM-4.5 | Use when the workload needs 128K context and structured outputs. | 2025-01 | 128K contextstructured outputs | Current |
| GLM-4.5-Air | Use when the workload needs 128K context and structured outputs. | 2025-01 | 128K contextstructured outputs | Current |
| GLM-4.6 | Use when the workload needs 198K context and structured outputs. | 2025-01 | 198K contextstructured outputs | Current |
| GLM-4.7 Flash | Use when the workload needs 198K context, reasoning, and JSON / Tool use. | 2025-01 | 198K contextreasoningJSON / Tool use | Current |
| GLM-4 9B | Use when the workload needs 128K context and 9B parameters. | 2024-06 | 128K context9B parameters | Current |
| GLM-4-Air | Use when the workload needs 128K context. | 2024-06 | 128K context | Current |
| GLM-4-Flash | Use when the workload needs 128K context. | 2024-06 | 128K context | Current |
| GLM-4-Flash-250414 | Use when the workload needs JSON / Tool use, structured outputs, and prompt caching. | Unknown release | JSON / Tool usestructured outputsprompt caching | Current |
| GLM-4-Voice | Use when the workload needs realtime voice, multimodal inputs, and audio. | Unknown release | realtime voicemultimodal inputsaudio | Current |
| GLM-4V 9B | Use when the workload needs 128K context, 9B parameters, and multimodal inputs. | 2024-06 | 128K context9B parametersmultimodal inputs | Current |
| GLM-4V-Flash | Use when the workload needs vision and multimodal inputs. | Unknown release | visionmultimodal inputs | Current |
Release Timeline
8 release groupsSpecifications(18 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Reasoning | JSON / Tool use | Structured Outputs | Code Exec |
|---|---|---|---|---|---|---|---|---|---|
| GLM-4.1V-Thinking-Flash | — | — | — | Yes | Yes | Yes | No | No | No |
| GLM-4.1V-Thinking-FlashX | — | — | — | Yes | Yes | Yes | No | No | No |
| GLM 4.7 | 2026-03 | 200K | 358B (32B active) | No | No | No | Yes | Yes | Yes |
| GLM 4.6V | 2026-02 | 128K | 106B (12B active) | Yes | Yes | No | Yes | No | No |
| GLM 4.5V | 2026-01 | 64K | 106B (12B active) | Yes | Yes | No | Yes | No | No |
| GLM-4 Code 9B | 2025-05 | 128K | 9B | No | No | No | No | No | No |
| GLM-4 32B | 2025-03 | 128K | 32B | No | No | No | No | Yes | No |
| GLM-4.5 | 2025-01 | 128K | 355B (32B active) | No | No | No | No | Yes | No |
| GLM-4.5-Air | 2025-01 | 128K | 106B (12B active) | No | No | No | No | Yes | No |
| GLM-4.6 | 2025-01 | 198K | 355B (32B active) | No | No | No | No | Yes | No |
| GLM-4.7 Flash | 2025-01 | 198K | 30B (3B active) | No | No | Yes | Yes | Yes | No |
| GLM-4 9B | 2024-06 | 128K | 9B | No | No | No | No | No | No |
| GLM-4-Air | 2024-06 | 128K | — | No | No | No | No | No | No |
| GLM-4-Flash | 2024-06 | 128K | — | No | No | No | No | No | No |
| GLM-4-Flash-250414 | — | — | — | No | No | No | Yes | Yes | No |
| GLM-4-Voice | — | — | — | No | Yes | No | No | No | No |
| GLM-4V 9B | 2024-06 | 128K | 9B | No | Yes | No | No | No | No |
| GLM-4V-Flash | — | — | — | Yes | Yes | No | No | No | No |
Available From(13 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| GLM-4V 9B | Replicate API | $0.05 | $0.25 | Serverless |
| GLM-4.7 Flash | Cloudflare Workers AI | $0.06 | $0.4 | Serverless |
| GLM-4.7 Flash | OpenRouter | $0.07 | $0.4 | Serverless |
| GLM-4.7 Flash | Vercel AI Gateway | $0.07 | $0.4 | Serverless |
| GLM-4.7 Flash | Novita AI | $0.07 | $0.4 | Serverless |
| GLM-4 32B | OpenRouter | $0.1 | $0.1 | Serverless |
| GLM-4 9B | AWS Bedrock | $0.1 | $0.1 | Serverless |
| GLM-4 9B | GCP Vertex AI | $0.1 | $0.1 | Serverless |
| GLM-4.5-Air | Novita AI | $0.13 | $0.85 | Serverless |
| GLM-4 9B | Bitdeer AI | $0.14 | $0.42 | Serverless |
| GLM-4 9B | Fireworks AI | $0.2 | $0.2 | Serverless |
| GLM-4.5-Air | OpenRouter | $0.2 | $1.1 | Serverless |
| GLM-4.5-Air | Vercel AI Gateway | $0.2 | $1.1 | Serverless |
| GLM-4.5-Air | Z.ai | $0.2 | $1.1 | Serverless |
| GLM 4.6V | Vercel AI Gateway | $0.3 | $0.9 | Serverless |
| GLM 4.6V | Novita AI | $0.3 | $0.9 | Serverless |
| GLM 4.6V | OpenRouter | $0.3 | $0.9 | Serverless |
| GLM 4.6V | Z.ai | $0.3 | $0.9 | Serverless |
| GLM-4.6 | Novita AI | $0.55 | $2.2 | Serverless |
| GLM 4.7 | Featherless | $0.55 | $2.2 | Serverless |
| GLM 4.7 | GCP Vertex AI | $0.6 | $2.2 | Serverless |
| GLM 4.7 | OpenRouter | $0.6 | $2.2 | Serverless |
| GLM-4.6 | OpenRouter | $0.6 | $2.2 | Serverless |
| GLM-4.5 | OpenRouter | $0.6 | $2.2 | Serverless |
| GLM 4.7 | Fireworks AI | $0.6 | $2.2 | Serverless |
| GLM-4.5 | Vercel AI Gateway | $0.6 | $2.2 | Serverless |
| GLM 4.5V | Vercel AI Gateway | $0.6 | $1.8 | Serverless |
| GLM-4.6 | Vercel AI Gateway | $0.6 | $2.2 | Serverless |
| GLM 4.7 | Novita AI | $0.6 | $2.2 | Serverless |
| GLM-4.5 | Novita AI | $0.6 | $2.2 | Serverless |
| GLM 4.5V | Novita AI | $0.6 | $1.8 | Serverless |
| GLM 4.5V | OpenRouter | $0.6 | $1.8 | Serverless |
| GLM 4.7 | Z.ai | $0.6 | $2.2 | Serverless |
| GLM-4.6 | Z.ai | $0.6 | $2.2 | Serverless |
| GLM-4.5 | Z.ai | $0.6 | $2.2 | Serverless |
| GLM 4.5V | Z.ai | $0.6 | $1.8 | Serverless |
| GLM-4.5 | Fireworks AI | $0.9 | $0.9 | Serverless |
| GLM-4.7 Flash | Fireworks AI | $0.9 | $0.9 | Serverless |
| GLM-4.5-Air | Fireworks AI | $0.9 | $0.9 | Serverless |
| GLM-4.6 | Fireworks AI | $0.9 | $0.9 | Serverless |
| GLM 4.7 | Vercel AI Gateway | $2.25 | $2.75 | Serverless |
Popular comparisons in this family
Models(18)
GLM-4.1V-Thinking-Flash
GLM-4.1V-Thinking-FlashX
GLM 4.7
GLM 4.6V
GLM 4.5V
GLM-4 Code 9B
GLM-4 32B
GLM-4.5
GLM-4.5-Air
GLM-4.6
GLM-4.7 Flash
GLM-4 9B
GLM-4-Air
GLM-4-Flash
GLM-4-Flash-250414
GLM-4-Voice
GLM-4V 9B
GLM-4V-Flash