DeepSeek-V3 vs GLM 4.7
DeepSeek vs Zhipu AI — benchmarks, pricing, and capabilities side by side.
- •GLM 4.7 has the higher intelligence index (89 vs 58.1)
- •DeepSeek-V3 is cheaper ($0.23 vs $0.40 per 1M input)
- •DeepSeek-V3 is faster
- •GLM 4.7 has a larger context window (203K)
| DeepSeek-V3 | GLM 4.7 | |
|---|---|---|
| Intelligence index | 58.1 | 89 |
| Developer | DeepSeek | Zhipu AI |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 131,072 tokens | 202,752 tokens |
| Input price | $0.23 / 1M | $0.40 / 1M |
| Output price | $0.91 / 1M | $1.75 / 1M |
| Speed | 100 tok/s | 98 tok/s |
| Released | December 26, 2024 | December 22, 2025 |
| Parameters | 671B total / 37B active (MoE) | — |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek-V3
GLM 4.7
AIME 2025
26
95
GPQA Diamond
59.1
85.9
Humanity’s Last Exam
3.6
25.1
LiveCodeBench
37.6
89.4
MMLU-Pro
75.9
85.6