GLM-4.5
| Released | 2025-07-28 |
|---|---|
| Price per 1M | $0.60 in · $2.20 out |
| Overall score | 46.4 · rank 147 |
| Context window | 131,072 tokens |
| Provider | Zhipu AI |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 46.4 | 147 | 8 |
| Overall: Coding | 46.7 | 137 | 3 |
| Overall: Reasoning | 45.6 | 157 | 4 |
| Overall: Agentic work | 46.7 | 139 | 3 |
| Overall: Frontend | 46.6 | 119 | 1 |
| Coding | 46.6 | 142 | 1 |
| Reasoning | 46.8 | 139 | 2 |
| Knowledge | 42.9 | 126 | 2 |
| General | 48.9 | 120 | 3 |
| Value vs price | −1.3 | 136 |
Evidence
Coding 46.6 ±5.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LiveCodeBench (Vals run) | 67.4% | Vals AI | 2026-09-01 |
Reasoning 46.8 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond (Vals run) | 72.2% | Vals AI | 2026-09-01 |
| Kagi Reasoning Benchmark | 57.9% | Kagi | 2026-11-11 |
Knowledge 42.9 ±1.3
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Humanity's Last Exam | 8.3% | Epoch AI | 2026-11-11 |
| MMLU-Pro (Vals run) | 81.2% | Vals AI | 2026-09-01 |
General 48.9 ±3.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena Text | 1,411 | LMArena | 2026-11-09 |
| EQ-Bench Creative Writing v3 | 1,344 | EQ-Bench | 2026-11-11 |
| EQ-Bench Longform Creative Writing | 55.5% | EQ-Bench | 2026-11-11 |
Names matched to this model
glm-4-5