Zhipu AI model product
7 is a coding‑centric model that thinks before acting, preserves its reasoning across turns, and lets you control thinking per request for speed or accuracy.
Updated Aug 12, 2026. Default version: GLM-4.7
Technical details for the model's default version.
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for GLM-4.7.
| Tau-bench | 87.4% | 02 | 6 | 80.0% | C | |
| BrowseComp-zh | 66.6% | 05 | 13 | 66.7% | C | |
| LiveCodeBench v6 | 84.9% | 10 | 53 | 82.7% | C | |
| IMO-AnswerBench | 82.0% | 12 | 19 | 38.9% | C | |
| Terminal-Bench | 33.3% | 16 | 25 | 37.5% | C | |
| AIME 2025 | 95.7% | 21 | 114 | 82.3% | C | |
| SWE-bench Multilingual | 66.7% | 23 | 34 | 33.3% | C | |
| MMLU-Pro | 84.3% | 26 | 129 | 80.5% | C | |
| Humanity's Last Exam | 42.8% | 30 | 93 | 68.5% | C | |
| BrowseComp | 52.0% | 40 | 58 | 31.6% | C | |
| SWE-Bench Verified | 73.8% | 40 | 105 | 62.5% | C | |
| Terminal-Bench 2.0 | 41.0% | 44 | 49 | 10.4% | C | |
| GPQA | 85.7% | 48 | 234 | 79.8% | C |
Preference and agent-evaluation results for the default version.
| webdev | overall | 51 | 1434.3 | 3,816 | N/A | |
| text | overall | 70 | 1435.4 | 12,163 | N/A | |
| text factuality | overall | 71 | 1443.3 | 11,743 | N/A | |
| text style control | overall | 76 | 1441.7 | 12,163 | N/A |
Provider-specific output speed and catalog latency for GLM-4.7. Runtime does not affect the capability score.
No 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.
Official vendor API pricing appears first, followed by individual provider offers.
| Zhipu AI Coding Plan | glm-4.7 | global | N/A | N/A | 204.8K | |
| Alibaba Coding Plan (China) | glm-4.7 | global | N/A | N/A | 202.8K | |
| Alibaba Coding Plan | glm-4.7 | global | N/A | N/A | 202.8K | |
| KUAE Cloud Coding Plan | GLM-4.7 | global | N/A | N/A | 204.8K | |
| Z.AI Coding Plan | glm-4.7 | global | N/A | N/A | 204.8K | |
| NanoGPT | zai-org/glm-4.7 | global | $0.20 | $0.80 | 200K | |
| Meganova | zai-org/GLM-4.7 | global | $0.20 | $0.80 | 202.8K | |
| CrofAI | glm-4.7 | global | $0.25 | $1.1 | 202.8K | |
| ZenMux | z-ai/glm-4.7 | global | $0.28 | $1.14 | 200K | |
| 302.AI | glm-4.7 | global | $0.286 | $1.14 | 204.8K | |
| LLM Gateway | glm-4.7 | global | $0.38 | $1.98 | 204.8K | |
| Ofox | z-ai/glm-4.7 | global | $0.40 | $2 | 204.8K | |
| OpenRouter | z-ai/glm-4.7 | global | $0.40 | $1.75 | 204.8K | |
| Deep Infra | zai-org/GLM-4.7 | global | $0.40 | $1.75 | 202.8K | |
| DInference | glm-4.7 | global | $0.45 | $1.65 | 200K | |
| CrossModel | z-ai/glm-4.7 | global | $0.47 | $2.16 | 200K | |
| Venice AI | zai-org-glm-4.7 | global | $0.55 | $2.65 | 198K | |
| Zhipu AI | glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Impossibl | zai/glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Baseten | zai-org/GLM-4.7 | global | $0.60 | $2.2 | 200K | |
| OpenCode Zen | glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Abacus | zai-org/GLM-4.7 | global | $0.60 | $2.2 | 204.8K | |
| OrcaRouter | z-ai/glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Vertex | zai-org/glm-4.7-maas | global | $0.60 | $2.2 | 200K | |
| Amazon Bedrock | zai.glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| TensorX | z-ai/glm-4.7 | global | $0.60 | $2.2 | 200K | |
| Kilo Gateway | z-ai/glm-4.7 | global | $0.60 | $2.2 | 202.8K | |
| Merge Gateway | zai/glm-4.7 | global | $0.60 | $2.2 | 200K | |
| Vercel AI Gateway | zai/glm-4.7 | global | $0.60 | $2.2 | 200K | |
| NovitaAI | zai-org/glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Hugging Face | zai-org/GLM-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Jiekou.AI | zai-org/glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Z.AI | glm-4.7 | global | $0.60 | $2.2 | 204.8K | |
| Berget.AI | zai-org/GLM-4.7 | global | $0.77 | $2.75 | 128K | |
| Cortecs | glm-4.7 | global | $0.78 | $2.79 | 202.8K | |
| Moark | GLM-4.7 | global | $3.5 | $14 | 204.8K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Available versions of this model. The score column identifies the version used in the overall ranking.
| GLM-4.7 | 59.7 | 358B | 204.8K | 131.1K | Yes | MIT |
Key information about GLM 4.7 and its available data.
7 is a coding‑centric model that thinks before acting, preserves its reasoning across turns, and lets you control thinking per request for speed or accuracy.
It upgrades agentic workflows with stronger multi‑step tool use, better terminal and multilingual coding, and a noticeable jump in UI output quality for modern, clean webpages and slides. ai API, and even run it locally with public weights on HuggingFace and ModelScope using vLLM or SGLang.
Data as of 2026-08-11.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Common questions about GLM 4.7.
GLM 4.7's default version was released on Dec 22, 2025.
GLM 4.7'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 $0.20 input and $0.80 output via NanoGPT.
GLM 4.7 was created by Zhipu AI.
The default version has a 204.8K token context window.
Yes. The default version is marked as open weight under MIT.
37 provider offerings are linked to the default version.
Nearby ranked alternatives include GPT-5.4-mini, Step 3.5 Flash, Gemma 4 31B.