GPT-4.1
Language modelPrevious generationCard summarizedSmartest non-reasoning model
Active (legacy)● Re-checked 12 Oct 2026 · 34 facts · 1 correction · 1 unconfirmedSource: GPT-4.1 Prompting Guide (OpenAI Cookbook) and GPT-4.1 model page; no system card published
Use it via API
Where to call it
- OpenAI API
gpt-4.1
Supports
- Image input
- Streaming
- Structured outputs
- Tool calling
- Web search
- Prompt caching
- Batch API
- Fine-tuning
Good for
- Instruction following and tool calling, with low latency
- Agentic coding: OpenAI's harness solved 55% of SWE-bench Verified
- Long-context work: document parsing, re-ranking, multi-hop reasoning over provided context
Overview
GPT-4.1
Smartest non-reasoning model
- Cost
- $2.00 in
- Context
- 1M
- Input / output
- $2.00 / $8.00
- Max output
- 32K
- Knows up to
- June 2024
- Released
- 14 Apr 2025
OpenAI published no system card for GPT-4.1. The model page calls it the "smartest non-reasoning model": strong at instruction following and tool calling, with a 1M-token context window and no reasoning step. OpenAI's prompting guide says it follows instructions more closely and literally than earlier models, so prompts may need migrating. OpenAI now recommends starting with GPT-5 for complex tasks. model page; Prompting Guide intro
gpt-4.1- Input / output per 1M
- $2.00 / $8.00
- Context window
- 1M
- Max output
- 32K
- Knows up to
- June 2024
- Reasoning
- None
- Weights
- Closed
About this card. A model card is the maker's own account of a model: what it's for, how it was trained and tested, and where it falls short. This page condenses OpenAI's GPT-4.1 Prompting Guide (OpenAI Cookbook) and GPT-4.1 model page; no system card published. Every claim carries its source page.