Phi-3-small-instruct (128k) vs GPT-5
Phi-3-small-instruct (128k) is cheaper on output tokens, while GPT-5 offers a larger context window. Choose Phi-3-small-instruct (128k) or GPT-5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Phi-3-small-instruct (128k) | GPT-5 |
|---|---|---|
| Provider | Azure Cognitive Services | Azure Cognitive Services |
| Input / 1M tokens | $0.15 | $1.25 |
| Output / 1M tokens | $0.60 | $10.00 |
| Context window | 128K | 272K |
| Parameters | — | — |
| Open weights | Yes | No |
| Released | Apr 2024 | Aug 2025 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.