GPT-4.1
GPT-4.1 is an OpenAI API model released April 14, 2025. Features a 1M token context window, 54.6% on SWE-bench Verified, and precise literal instruction following. Designed for developers building agentic coding workflows.
Technical specification
Modalities
Capabilities
12Reasoning★
Reasoning
Multi-step reasoning★
Reasoning
Long context★
Reasoning
Coding★
Coding
Function Calling
Planning
Structured output★
Structured gen.
Image understanding★
Vision
Chart understanding
Vision
OCR★
Vision
Multilingual★
Language
Planning★
Planning
Streaming output
Reasoning
Applications
$2.00 per million input tokens, $8.00 per million output tokens. Cached input: $0.50/MTok (75% discount). Batch API: 50% discount ($1.00/$4.00). Fine-tuned inference: ~$3/$12 per MTok. No price premium for long context up to 1M tokens. Price ~26% lower than GPT-4o.
INPUT
$2.0000 / per 1M tokens
OUTPUT
$8.0000 / per 1M tokens
CACHE
$0.5000 / per 1M tokens
TOTAL
for 10K tokens
Standard text tokens. Cached input = $0.50/MTok (75% discount). No surcharge for long context up to 1M tokens.
50% discount via the asynchronous Batch API (processing within 24h). Available for GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano.
Estimated inference pricing after fine-tuning. Fine-tuning training: ~$3.00/MTok. Available via OpenAI API and Azure AI Foundry.
Security and enterprise
OpenAI publishes security and enterprise documentation for its platform, covering the API and ChatGPT Enterprise/Business/Edu offerings. For GPT-4.1, security information is platform-level rather than a dedicated per-model safety sheet. Publicly documented aspects include data encryption, access controls, compliance certifications, and policies on the use of customer data for model training.
Technical information
Security information for GPT-4.1 should be treated as pertaining to the OpenAI API environment and enterprise products, rather than as a model-specific security specification. In practice, this is the appropriate approach for an AI systems catalog.