GLM-4.5
Zero-eval
#1MATH-500
#1BFCL-v3
#1AA-Index
+4 more
by Zhipu AI
+
+
+
+
About
GLM-4.5 is a language model developed by Zhipu AI. It achieves strong performance with an average score of 64.0% across 14 benchmarks. It excels particularly in MATH-500 (98.2%), AIME 2024 (91.0%), MMLU-Pro (84.6%). It supports a 262K token context window for handling large documents. The model is available through 3 API providers. It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Zhipu AI's latest advancement in AI technology.
+
+
+
+
Pricing Range
Input (per 1M)$0.40 -$0.60
Output (per 1M)$1.60 -$2.20
Providers3
+
+
+
+
Timeline
AnnouncedJul 28, 2025
ReleasedJul 28, 2025
+
+
+
+
Specifications
Training Tokens23.0T
+
+
+
+
License & Family
License
MIT
Performance Overview
Performance metrics and category breakdown
Overall Performance
14 benchmarks
Average Score
64.0%
Best Score
98.2%
High Performers (80%+)
3Performance Metrics
Max Context Window
262.1KAvg Throughput
85.0 tok/sAvg Latency
1ms+
+
+
+
All Benchmark Results for GLM-4.5
Complete list of benchmark scores with detailed information
| MATH-500 | text | 0.98 | 98.2% | Self-reported | |
| AIME 2024 | text | 0.91 | 91.0% | Self-reported | |
| MMLU-Pro | text | 0.85 | 84.6% | Self-reported | |
| TAU-bench Retail | text | 0.80 | 79.7% | Self-reported | |
| GPQA | text | 0.79 | 79.1% | Self-reported | |
| BFCL-v3 | text | 0.78 | 77.8% | Self-reported | |
| LiveCodeBench | text | 0.73 | 72.9% | Self-reported | |
| AA-Index | text | 0.68 | 67.7% | Self-reported | |
| SWE-Bench Verified | text | 0.64 | 64.2% | Self-reported | |
| TAU-bench Airline | text | 0.60 | 60.4% | Self-reported |
Showing 1 to 10 of 14 benchmarks