Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | GPT-3.5 Turbo | DeepSeek V3.1 Terminus (Non-reasoning) |
|---|---|---|
| Input ($/M tokens) | $0.5 | $0.335 |
| Output ($/M tokens) | $1.5 | $1.5 |
Data from Artificial Analysis API — 12 benchmarks
DeepSeek V3.1 Terminus (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.63/M tokens vs $0.75/M for GPT-3.5 Turbo.
DeepSeek V3.1 Terminus (Non-reasoning) wins 11 out of 12 benchmarks compared to 1 for GPT-3.5 Turbo. See the detailed benchmark chart above for per-category results.
GPT-3.5 Turbo generates tokens faster at 100 tok/s vs 0 tok/s. However, DeepSeek V3.1 Terminus (Non-reasoning) has lower time-to-first-token (0.00s vs 0.43s).
Choose based on your priorities: DeepSeek V3.1 Terminus (Non-reasoning) for lower cost, DeepSeek V3.1 Terminus (Non-reasoning) for stronger benchmark performance, and GPT-3.5 Turbo for faster generation. For latency-sensitive apps, check the TTFT comparison above.