According to UsagePricing's tracker, davinci-002 costs $2.00 per 1M input tokens and $2.00 per 1M output tokens as of August 2026. The model supports a 16,384-token context window.
Legacy GPT-3 base model on the Completions API — flat input/output pricing.
Token Costs (per 1M)
Input Price
$2.00
Output Price
$2.00
Technical Specs
Price History
| Date | Input Price | Output Price |
|---|---|---|
| 8/31/2026 | $2.00 | $2.00 |
Compare davinci-002 With Other Models
Building with davinci-002?
Estimate your total monthly bill based on your expected traffic and usage patterns. Our advanced calculator handles complex scenarios like RAG and agentic workflows.
Frequently Asked Questions
How much does davinci-002 cost per 1M tokens? ▼
davinci-002 costs $2 per 1M input tokens and $2 per 1M output tokens as of August 2026.
What is davinci-002's context window? ▼
davinci-002 has a context window of 16,384 tokens, meaning it can process approximately 12,288 words in a single request.
Who makes davinci-002? ▼
davinci-002 is made by OpenAI. You can access it via the OpenAI API.