Best LLMs for Science — Z.ai
The best Z.ai model for science is GLM 5.2, scoring 89.5% on GPQA Diamond — graduate-level physics, chemistry and biology. GLM 5.1 (86.8%) and GLM 4.7 (85.9%) round out the top three.
- 1
glm-5.2ReasoningToolsJSON51.1 intel · $0.760/M · 179 t/s89.5%GPQA - 2glm-5.1ReasoningToolsJSON40.2 intel · $0.966/M · 205K ctx86.8%GPQA
- 3glm-4.7ReasoningToolsJSON33.7 intel · $0.400/M · 496 t/s85.9%GPQA
- 4glm-5-turboReasoningToolsJSON38.1 intel · $1.20/M · 203K ctx84.7%GPQA
- 5glm-5ReasoningToolsJSON39.5 intel · $0.950/M · 95 t/s82.0%GPQA
- 6glm-5v-turboReasoningToolsJSON+134.5 intel · $1.20/M · 203K ctx80.9%GPQA
- 7glm-4.5ReasoningToolsJSON19.5 intel · $0.600/M · 33 t/s78.2%GPQA
- 8glm-4.5-airReasoningTools16.5 intel · $0.130/M · 42 t/s73.3%GPQA
- 9glm-4.6ReasoningToolsJSON23.0 intel · $0.500/M · 38 t/s63.2%GPQA
- 10glm-4.7-flashReasoningToolsJSON22.9 intel · $0.060/M · 43 t/s58.1%GPQA
- 11glm-4.5vReasoningToolsJSON+17.0 intel · $0.600/M · 39 t/s57.3%GPQA
- 12glm-4.6vReasoningToolsJSON+111.0 intel · $0.300/M · 33 t/s56.6%GPQA
How this is ranked
AI models ranked by GPQA Diamond — graduate-level physics, chemistry and biology questions written to be un-Googleable, so the score reflects reasoning from knowledge rather than retrieval. The best large language models for scientific research and hard-science work.
Frequently asked
What is the best Z.ai model for science?
The best Z.ai model for science is GLM 5.2, scoring 89.5% on GPQA Diamond — graduate-level physics, chemistry and biology. GLM 5.1 (86.8%) and GLM 4.7 (85.9%) round out the top three.
What's a good alternative to GLM 5.2?
GLM 5.1 (86.8%) is the closest alternative on this metric, followed by GLM 4.7 (85.9%). See the full ranking above for the tradeoffs.
How many Z.ai models are there?
modelgrep tracks 12 Z.ai models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by GLM 5.2. 12 of them qualify for this ranking.