DeepSeek V3.2 vs GLM 4.7
Head to head on 22 shared benchmarks, with price, context window, and release dates.
Benchmark scores
41evals · sort by any column, ⤢ to expand fullscreen
| Model | ||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GLM 4.7 Zai | 48.0% | 46.4% | 1411elo | 66 | 2.4% | 66.4% | 6.4% | 54.6% | 60.8% | 73.1% | 55.2% | 35.7% | 65.2% | 76.0% | 59.7% | 56.2% | 79.4% | 6.0% | 35.4% | 68.8% | 30.3% | 94.2% | - | - | - | - | 46.7% | 28.6% | 90.0% | - | - | 33.3% | - | 32.8% | 33.3% | 68.6% | 100.0% | - | - | - | - | 51.9% |
| DeepSeek V3.2 DeepSeek | 59.0% | 47.9% | 1511elo | 66.8 | 22.1% | 75.1% | 11.2% | 49.0% | 65.9% | 64.6% | 50.0% | 48.2% | 70.4% | 85.0% | 77.2% | 59.3% | 83.7% | 8.0% | 38.7% | 68.2% | 32.6% | 78.9% | 74.7% | 74.2% | 0.0% | 16.0% | - | - | - | 31.3% | 68.0% | - | 57.5% | - | - | - | - | 59.0% | 70.0% | 65 | 5.1% | 51.7% |
2 / 2 models
| Attribute | DeepSeek V3.2 DeepSeek | GLM 4.7 Zai |
|---|---|---|
| Description | DeepSeek V3.2 is an AI model from DeepSeek, released with open weights. | GLM-4.7 is an AI model from Zai, released with open weights. |
| Family | Deepseek | Glm |
| Params | - | - |
| Open weights | open | open |
| License | mit | mit |
| Released | 1 Dec 2025 | 22 Dec 2025 |
| Context | 163,840 | 204,800 |
| Input $/1M | $0.28 | $0.60 |
| Output $/1M | $0.42 | $2.20 |
| Cache read $/1M | $0.130 | $0.080 |
| Cache write $/1M | - | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | text | text |
| Providers | - | - |
| Publisher | DeepSeek | Zai |
| Reported scores | Aider Polyglot Benchmark74.2% (Aider) MathArena57.47% GPQA Diamond75.1% (AA) Humanity's Last Exam (HLE)11.2% (AA) |
FAQ
- Is DeepSeek V3.2 or GLM 4.7 better?
- Across the 22 benchmarks both models report on Sophon, DeepSeek V3.2 leads on 17 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do DeepSeek V3.2 and GLM 4.7 compare on GPQA Diamond?
- DeepSeek V3.2 scores 75.1% and GLM 4.7 scores 66.4% on GPQA Diamond.
- Which is cheaper, DeepSeek V3.2 or GLM 4.7?
- DeepSeek V3.2 is cheaper at $0.42 per million output tokens against $2.20 for GLM 4.7 - about 5.2x.
- When were DeepSeek V3.2 and GLM 4.7 released?
- DeepSeek V3.2 was released 1 Dec 2025 by DeepSeek. GLM 4.7 was released 22 Dec 2025 by Zai.
Comparing something else? Build your own side-by-side.