MODEL PRICE SHEET
GPT-4.1 API Pricing
Standard text-token prices, source details, and a monthly cost estimate.
STANDARD RATES
What does this model cost?
USD per 1M tokens. Standard text-token rate. Batch API has separate pricing.
INPUT$2.00per 1M tokens
OUTPUT$8.00per 1M tokens
CACHED$0.50per 1M tokens
Batch pricing: see the official provider page for current rates. Source checked: 2026-10-03 05:39 UTC. Official pricing source ↗
Monthly cost examples
Standard rates, with no cached input.
Customer support bot10M input + 2M output / monthly estimate
$36.00Coding assistant4M input + 2M output / monthly estimate
$24.00Document Q&A8M input + 1M output / monthly estimate
$24.00Compare similar models
| Provider / Model | Context | Input / 1M | Output / 1M | Cached / 1M | Coding | |
|---|---|---|---|---|---|---|
Claude Sonnet 4Anthropic (Claude) · Flagship | 200K | $3.00 | $15.00 | $0.30 | ● Yes | Details ↗ |
GPT-4.1 MiniOpenAI · Efficient | 1M | $0.40 | $1.60 | $0.10 | ● Yes | Details ↗ |
Gemini 2.5 FlashGoogle (Gemini) · Efficient | 1M | $0.30 | $2.50 | $0.03 | — | Details ↗ |
Frequently asked questions
How is this model billed?
Standard API usage is billed by input and output token volume.
Does caching reduce cost?
Eligible cached input uses the displayed cached rate. Provider rules determine eligibility. $0.50 per 1M tokens.
Are there extra fees?
Tools, grounding, storage, audio, and taxes may add charges. Review the official source. Official pricing source ↗