GLM 4.7 vs Llama 3.1 Nemotron Ultra 253B v1
Zhipu AI vs NVIDIA — benchmarks, pricing, and capabilities side by side.
- •GLM 4.7 has the higher intelligence index (89 vs 78.3)
- •GLM 4.7 is cheaper ($0.40 vs $0.60 per 1M input)
- •GLM 4.7 is faster
| GLM 4.7 | Llama 3.1 Nemotron Ultra 253B v1 | |
|---|---|---|
| Intelligence index | 89 | 78.3 |
| Developer | Zhipu AI | NVIDIA |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 202,752 tokens | — |
| Input price | $0.40 / 1M | $0.60 / 1M |
| Output price | $1.75 / 1M | $1.80 / 1M |
| Speed | 98 tok/s | 42 tok/s |
| Released | December 22, 2025 | April 7, 2025 |
| Parameters | — | 253000000000 |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
GLM 4.7
Llama 3.1 Nemotron Ultra 253B v1
AIME 2025
95
72.5
GPQA Diamond
85.9
76
Humanity’s Last Exam
25.1
8.1
LiveCodeBench
89.4
66.3
MMLU-Pro
85.6
82.5