Updated March 26, 2026· Based on independent benchmark data
GPT-5.2 (xhigh) leads in intelligence with a score of 51.3 vs 48.4. For speed, Gemini 3 Pro Preview (high) wins at 115 tok/s vs 70 tok/s.
| Metric | GPT-5.2 (xhigh) | Gemini 3 Pro Preview (high) |
|---|---|---|
| Intelligence Score | 51.3 | 48.4 |
| Coding Score | 48.7 | 46.5 |
| Math Score | 99.0 | 95.7 |
| Speed (tok/s) | 70 tok/s | 115 tok/s |
| Latency (TTFT) | 75.24s | 36.25s |
| Input Price / 1M tokens | $1.75 | $2.00 |
| Output Price / 1M tokens | $14 | $12 |
| Context Window | N/A |
GPT-5.2 (xhigh) outperforms Gemini 3 Pro Preview (high) on the intelligence index with a score of 51.3 compared to 48.4. For coding tasks, GPT-5.2 (xhigh) has the edge with a coding score of 48.7 vs 46.5. In mathematical reasoning, GPT-5.2 (xhigh) leads with 99.0 compared to Gemini 3 Pro Preview (high)'s 95.7.
Gemini 3 Pro Preview (high) generates output significantly faster at 115 tok/s compared to GPT-5.2 (xhigh)'s 70 tok/s, making it 1.7x faster for streaming responses. Time to first token is 36.25s for Gemini 3 Pro Preview (high) vs 75.24s for GPT-5.2 (xhigh), which affects perceived responsiveness in interactive applications.
GPT-5.2 (xhigh) is more affordable at $1.75/1M input tokens ($14/1M output), while Gemini 3 Pro Preview (high) costs $2.00/1M input ($12/1M output). For a typical workload of 100 requests per day at 2,000 tokens each, GPT-5.2 (xhigh) would cost approximately $10.50/month vs $12.00/month for Gemini 3 Pro Preview (high) in input costs alone.
Choose GPT-5.2 (xhigh) when you need higher intelligence (51.3), stronger coding performance (48.7). Choose Gemini 3 Pro Preview (high) when you need faster output (115 tok/s).
GPT-5.2 (xhigh) scores higher on coding benchmarks (48.7 vs 46.5), making it the better choice for programming tasks.
GPT-5.2 (xhigh) is cheaper at $1.75/1M input tokens vs $2.00/1M for Gemini 3 Pro Preview (high).
Gemini 3 Pro Preview (high) is faster, producing output at 115 tok/s compared to GPT-5.2 (xhigh)'s 70 tok/s.
No, GPT-5.2 (xhigh) does not support image input. Neither model supports image input.
Data last synced: March 26, 2026
| N/A |
| Max Output Tokens | N/A | N/A |
| Input Modalities | Text | Text |
| Output Modalities | Text | Text |
| Free Tier | No | No |
It depends on your priorities. GPT-5.2 (xhigh) scores higher on intelligence (51.3), but Gemini 3 Pro Preview (high) may be better for specific use cases like budget-conscious projects or speed-critical applications.