Compare models
Comparing 2models. Drop the URL into a doc — it's permalinked.
| Field | google/gemini-2.5-flash | openai/gpt-4o |
|---|---|---|
| Provider | openai | |
| Model ID | gemini-2.5-flash | gpt-4o |
| Context | 1.0M | 128K |
| Max output | 8K | 16K |
| Input / 1M | $0.30 | $2.50 |
| Output / 1M | $2.50 | $10.00 |
| Cached input / 1M | $0.07 | $1.25 |
| Avg cost / 1M | $1.40 | $6.25 |
| Speed | 250 t/s | 110 t/s |
| Quality index | 70.0 | 71.0 |
| MMLU | 85.3 | 88.7 |
| GPQA | 65.2 | 53.6 |
| HumanEval | 88.1 | 90.2 |
| MATH | 88.0 | 76.6 |
| SWE-bench | — | — |
| Arena Elo | — | 1338 |
| Tools | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Thinking | ✓ | — |
| Streaming | ✓ | ✓ |
| JSON mode | ✓ | ✓ |
| Structured output | ✓ | ✓ |
| Prompt cache | — | — |
Same data, in your terminal: relay models compare gemini-flash 4o