GPT-3.5 Turbo pricing
Source: litellm · updated 2026-10-28
Short answer: GPT-3.5 Turbo costs $0.500 per million input tokens and $1.50 per million output tokens. That is a 3:1 blended rate of $0.750 per million tokens, ranking it #383 of 713 tracked models by input price.
Input / 1M
$0.500
Output / 1M
$1.50
Output ÷ input
3.00×
Context window
16,385
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $0.500 |
| Output, per 1M tokens | $1.50 |
| Output ÷ input multiple | 3.00× |
| 3:1 blended, per 1M tokens | $0.750 |
| Context window | 16,385 tokens |
| Vendor | vercel_ai_gateway |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total |
|---|---|---|---|
| 1M in / 250K out | $0.500 | $0.375 | $0.875 |
| 10M in / 2.5M out | $5.00 | $3.75 | $8.75 |
| 100M in / 25M out | $50.00 | $37.50 | $87.50 |
Other vercel_ai_gateway models
| Model | Input | Output |
|---|---|---|
| Ministral 3 B | $0.040 | $0.040 |
| Llama 3.8 B | $0.050 | $0.080 |
| Devstral Small | $0.070 | $0.280 |
| Qwen 3 14 B | $0.080 | $0.240 |
| Qwen 3 30 B | $0.100 | $0.300 |
Compare against
Cheaper alternatives
Next cheaper option: Qwen 2.5 Coder 32 B Instruct at $0.660 per 1M input tokens. Next step up: Minicpm V4.5 at $0.658 per 1M input tokens.
Spotted a price that is out of date? Submit a correction.