Mk1.5 vs GPT-3.5 Turbo
GPT-3.5 Turbo is cheaper on output tokens, while Mk1.5 offers a larger context window. Choose Mk1.5 or GPT-3.5 Turbo based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Mk1.5 | GPT-3.5 Turbo |
|---|---|---|
| Provider | AI/ML API | OpenAI |
| Input / 1M tokens | $0.16 | $0.50 |
| Output / 1M tokens | $1.59 | $1.50 |
| Context window | 37K | 16K |
| Parameters | — | 20B |
| Open weights | No | No |
| Released | Sep 2026 | Mar 2023 |
| LMArena (Chatbot Arena) Elo score | — | 1125 |
| GPQA Diamond score | — | 28.0 |
| Epoch Capabilities Index score | — | 119 |
FAQ
Is Mk1.5 or GPT-3.5 Turbo cheaper?
GPT-3.5 Turbo is cheaper on output tokens ($1.50 vs $1.59 per 1M).
Which has the larger context window, Mk1.5 or GPT-3.5 Turbo?
Mk1.5 has the larger context window (37K tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.