GLM-4.5
Summary
GLM-4.5 is an open-weight language model from Zhipu AI (Z.ai), released on 28 Jul 2025. It accepts text and generates text, with a 131K-token context window and up to 98K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.60 per million input tokens and $2.20 per million output tokens (Z.ai API), across 4 providers we track. It is scheduled for retirement on 31 Dec 2026.
Hybrid-reasoning GLM release that made the 4.5 line broadly useful
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Zhipu AI (Z.ai) |
|---|---|
| Release date | 28 Jul 2025 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | mit [Hugging Face] |
| Parameters | 358B [Hugging Face] |
| Context window | 131K (131,072 tokens) [models.dev] |
| Max output | 98K tokens [models.dev] |
| Knowledge cutoff | Apr 2025 [models.dev] |
| Input | Text [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Retirement date | 31 Dec 2026 [OpenRouter] |
| Hugging Face | zai-org/GLM-4.5 [Hugging Face] |
GLM-4.5 API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Z.ai API glm-4.5 | $0.60 | $2.20 | $0.11 | 131K | models.dev, 18 Nov 2026 |
| Hugging Face Inference Providers zai-org/GLM-4.5 | $0.60 | $2.20 | – | 131K | models.dev, 18 Nov 2026 |
| OpenRouter z-ai/glm-4.5 | $0.60 | $2.20 | $0.11 | 131K | OpenRouter, 18 Nov 2026 |
| Vercel AI Gateway zai/glm-4.5 | $0.60 | $2.20 | $0.11 | 128K | models.dev, 18 Nov 2026 |
Running GLM-4.5 locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 217 GB | 221 GB | 230 GB |
| Q8_0 (8-bit) | 381 GB | 385 GB | 394 GB |
| FP16 / BF16 (16-bit) | 717 GB | 721 GB | 730 GB |
At 4-bit with an 8K conversation, GLM-4.5 needs about 221 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of GLM-4.5?
GLM-4.5 has a context window of 131,072 tokens (131K) and can generate up to 98,304 tokens in a single response, according to models.dev.
How much does the GLM-4.5 API cost?
Through Z.ai API, GLM-4.5 costs $0.60 per million input tokens and $2.20 per million output tokens. Prices last verified 18 Nov 2026.
Is GLM-4.5 open source?
GLM-4.5 is an open-weight model: its weights are published on Hugging Face as zai-org/GLM-4.5 under the mit license. Check the license terms before commercial use.
When was GLM-4.5 released?
Zhipu AI (Z.ai) released GLM-4.5 on 28 Jul 2025, according to models.dev.
Which providers offer GLM-4.5?
We track GLM-4.5 on 4 providers: Z.ai API, Hugging Face Inference Providers, OpenRouter, Vercel AI Gateway.
How much VRAM does GLM-4.5 need?
About 221 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 385 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Kimi K2 Instruct | Moonshot AI | 11 Jul 2025 | 131K | $0.57 | $2.30 | Open |
| Kimi-K2-Instruct | Moonshot AI | 11 Jul 2025 | 131K | $0.57 | $2.30 | Open |
| Hermes-4-405B | Nousresearch | 26 Aug 2025 | 131K | $1 | $3 | Open |
| DeepSeek-R1-0528 | DeepSeek | 28 May 2025 | 164K | $0.50 | $2.15 | Open |
| Magistral Small 1.2 | Mistral AI | 18 Sept 2025 | 128K | $0.50 | $1.50 | Open |
| Qwen3 VL 235B A22B Thinking | Alibaba (Qwen) | 23 Sept 2025 | 131K | $0.98 | $3.95 | Open |
Change history
- Now listed on Hugging Face Inference Providers
- No longer listed on Hugging Face Inference Providers
- No longer listed on novita-ai
Sources
- https://openrouter.ai/z-ai/glm-4.5
- https://github.com/sst/models.dev/blob/dev/providers/zai/models/glm-4.5.toml
- https://huggingface.co/zai-org/GLM-4.5
- https://github.com/sst/models.dev/blob/dev/providers/huggingface/models/zai-org/GLM-4.5.toml
- https://github.com/sst/models.dev/blob/dev/providers/vercel/models/zai/glm-4.5.toml