Zhipu AI model product
GLM 4.5
5 is an Agentic, Reasoning, and Coding (ARC) foundation model designed for intelligent agents, featuring 355 billion total parameters with 32 billion active parameters using MoE architecture.
Updated Aug 12, 2026. Default version: GLM-4.5
LLMBoard score49.3GLM-4.5
Coverage80%14 benchmark families
Context window131.1KTokens
Official input price$0.60Zhipu AI API
GLM 4.5 Specifications
Technical details for the model's default version.
- Version
- GLM-4.5
- Released
- Jul 28, 2025
- Knowledge cutoff
- Apr 1, 2025
- Parameters
- 355B
- Context window
- 131.1K
- Max output
- 98.3K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- MIT
GLM 4.5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GLM 4.5 Benchmark Results
Benchmark scores for GLM-4.5.
| AA-Index | 67.7% | 01 | 3 | 100.0% | C | |
| BFCL-v3 | 77.8% | 01 | 19 | 100.0% | C | |
| MATH-500 | 98.2% | 03 | 32 | 93.5% | C | |
| TAU-bench Airline | 60.4% | 04 | 23 | 86.4% | C | |
| TAU-bench Retail | 79.7% | 06 | 25 | 79.2% | C | |
| AIME 2024 | 91.0% | 08 | 53 | 86.5% | C | |
| SciCode | 41.7% | 12 | 19 | 38.9% | C | |
| Terminal-Bench | 37.5% | 12 | 25 | 54.2% | C | |
| LiveCodeBench | 72.9% | 18 | 73 | 76.4% | C | |
| MMLU-Pro | 84.6% | 23 | 129 | 82.8% | C | |
| BrowseComp | 26.4% | 56 | 58 | 3.5% | C | |
| Humanity's Last Exam | 14.4% | 74 | 93 | 20.6% | C | |
| SWE-Bench Verified | 64.2% | 75 | 105 | 28.9% | C | |
| GPQA | 79.1% | 91 | 234 | 61.4% | C |
GLM 4.5 Arena Results
Preference and agent-evaluation results for the default version.
| text | overall | 75 | 1429.4 | 24,423 | N/A | |
| text style control | overall | 126 | 1411.2 | 24,423 | N/A |
GLM 4.5 Runtime Performance
Provider-specific output speed and catalog latency for GLM-4.5. Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GLM 4.5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.60 input, $2.2 output per 1M
- Official provider
- Zhipu AI
- Lowest third-party
- From $0.286 input, $1.14 output per 1M via 302.AI
- Tracked offerings
- 18
| ModelScope | ZhipuAI/GLM-4.5 | global | N/A | N/A | 131.1K | |
| 302.AI | glm-4.5 | global | $0.286 | $1.14 | 131.1K | |
| NanoGPT | zai-org/glm-4.5 | global | $0.30 | $1.3 | 128K | |
| ZenMux | z-ai/glm-4.5 | global | $0.35 | $1.54 | 128K | |
| Zhipu AI | glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Impossibl | zai/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| LLM Gateway | glm-4.5 | global | $0.60 | $2.2 | 131K | |
| Abacus | zai-org/GLM-4.5 | global | $0.60 | $2.2 | 131.1K | |
| OrcaRouter | z-ai/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Kilo Gateway | z-ai/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Merge Gateway | zai/glm-4.5 | global | $0.60 | $2.2 | 128K | |
| Vercel AI Gateway | zai/glm-4.5 | global | $0.60 | $2.2 | 128K | |
| NovitaAI | zai-org/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| OpenRouter | z-ai/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Hugging Face | zai-org/GLM-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Jiekou.AI | zai-org/glm-4.5 | global | $0.60 | $2.2 | 131.1K | |
| Z.AI | glm-4.5 | global | $0.60 | $2.2 | 131.1K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GLM 4.5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| GLM-4.5 | 49.3 | 355B | 131.1K | 98.3K | Yes | MIT |
What is GLM 4.5?
Key information about GLM 4.5 and its available data.
5 is an Agentic, Reasoning, and Coding (ARC) foundation model designed for intelligent agents, featuring 355 billion total parameters with 32 billion active parameters using MoE architecture.
Trained on 23T tokens through multi-stage training, it is a hybrid reasoning model that provides two modes: thinking mode for complex reasoning and tool usage, and non-thinking mode for immediate responses. The model unifies agentic, reasoning, and coding capabilities with 128K context length support.
2 across 12 industry-standard benchmarks, placing 3rd among all proprietary and open-source models. Released under MIT open-source license allowing commercial use and secondary development.
Data as of 2026-08-11.
GLM 4.5 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to GLM 4.5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ZA
GLM 4.6
Zhipu AI
ZA
GLM 5V Turbo
Zhipu AI
ZA
GLM 4.5 Air
Zhipu AI
ZA
GLM 4.7
Zhipu AI
ZA
GLM 4.7 Flash
Zhipu AI
ZA
GLM 5
Zhipu AI
FAQ
Common questions about GLM 4.5.
When was GLM 4.5 released?
GLM 4.5's default version was released on Jul 28, 2025.
How much does GLM 4.5 cost?
GLM 4.5's official API price is $0.60 per million input tokens and $2.2 per million output tokens via Zhipu AI. The lowest tracked third-party offer starts at $0.286 input and $1.14 output via 302.AI.
Who created GLM 4.5?
GLM 4.5 was created by Zhipu AI.
What is the context window for GLM 4.5?
The default version has a 131.1K token context window.
Is GLM 4.5 open weight?
Yes. The default version is marked as open weight under MIT.
How many API providers offer GLM 4.5?
18 provider offerings are linked to the default version.
What models should I compare GLM 4.5 with?
Nearby ranked alternatives include Gemini 2.5 Pro, GPT-5, GPT-5.1-Codex.