Compare models
Comparing 2models. Drop the URL into a doc — it's permalinked.
| Field | nvidia/llama-3.1-nemotron-70b-instruct | openai/gpt-4o-mini |
|---|---|---|
| Provider | nvidia | openai |
| Model ID | llama-3.1-nemotron-70b-instruct | gpt-4o-mini |
| Context | 131K | 128K |
| Max output | — | 16K |
| Input / 1M | $1.20 | $0.15 |
| Output / 1M | $1.20 | $0.60 |
| Cached input / 1M | — | $0.07 |
| Avg cost / 1M | $1.20 | $0.38 |
| Speed | — | 145 t/s |
| Quality index | — | 60.0 |
| MMLU | — | 82.0 |
| GPQA | — | 40.2 |
| HumanEval | — | 87.2 |
| MATH | — | 70.2 |
| SWE-bench | — | — |
| Arena Elo | — | — |
| Tools | ✓ | ✓ |
| Vision | — | ✓ |
| Thinking | — | — |
| Streaming | — | ✓ |
| JSON mode | ✓ | ✓ |
| Structured output | — | ✓ |
| Prompt cache | — | — |
Same data, in your terminal: relay models compare nvidia/llama-3.1-nemotron-70b-instruct 4o-mini