GPT-5.2 vs GLM-4.5-Air
OpenAI GPT-5.2 | Z.ai GLM-4.5-Air | |
|---|---|---|
| Overview | ||
| Company | OpenAI | Z.ai |
| Release date | Dec 11 2025 | Jul 28 2025 |
| Access | Proprietary | Open Weight |
| Specifications | ||
Parameters | — | 106B |
| Benchmarks | ||
Nonsense detection BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 38% | — |
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 80%Best | 59.8% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 92.4% | — |
| Timeline | ||
| Release gap | GLM-4.5-Air shipped 136 days before GPT-5.2 | |
Which is better: GPT-5.2 or GLM-4.5-Air?
GPT-5.2 leads GLM-4.5-Air on 1 of the 1 benchmark they both report (SWE-Bench Verified). GLM-4.5-Air shipped 136 days before GPT-5.2, so benchmark comparisons should account for the intervening progress.
GPT-5.2 is proprietary, while GLM-4.5-Air is open weight.
On SWE-Bench Verified, GPT-5.2 leads at 80% vs GLM-4.5-Air at 59.8%.
Frequently asked questions
GPT-5.2 was released by OpenAI on Dec 11 2025.
GLM-4.5-Air was released by Z.ai on Jul 28 2025.
GPT-5.2 leads on SWE-Bench Verified — GPT-5.2 80% vs GLM-4.5-Air 59.8%.
GPT-5.2 is a proprietary model released by OpenAI. GLM-4.5-Air is an open weight model released by Z.ai.
Other comparisons
GPT-5.2 vs Claude Fable 5GLM-4.5-Air vs Claude Fable 5GPT-5.2 vs Gemini OmniGLM-4.5-Air vs Gemini OmniGPT-5.2 vs Muse SparkGLM-4.5-Air vs Muse SparkGPT-5.2 vs Grok 4.3 BetaGLM-4.5-Air vs Grok 4.3 BetaGPT-5.2 vs DeepSeek-V4-ProGLM-4.5-Air vs DeepSeek-V4-ProGPT-5.2 vs Mistral Medium 3.5GLM-4.5-Air vs Mistral Medium 3.5