GPT-4.1
OpenAI GPT-4.1: a non-reasoning model with a 1M context window, predicted outputs, fine-tuning support, and API-only access for developer workflows.
Model Overview
Capabilities, design details, and architectural traits
GPT-4.1 - Non-Reasoning GPT, API-Only
GPT-4.1 is part of the GPT series and produces responses without a reasoning step, making it suited to tasks that need low latency, long context, and precise instruction execution. It is available via the API only - it is not in ChatGPT.
| Trait | Detail |
|---|---|
| Context window | 1,047,576 tokens (1M) with documented improved long-context comprehension, not just capacity |
| No reasoning step | Produces output without a chain-of-thought reasoning pass, delivering low latency compared to o-series models |
| API-only distribution | Explicitly not available in ChatGPT; accessible only through the API |
| Predicted outputs | Supported - allows supplying an expected output to speed up responses when the output is mostly known; not available on o3 or o4-mini |
| Fine-tuning | Supported via supervised fine-tuning and direct preference optimization |
| Extraneous code edits | Documented reduction from 9% with GPT 4o to 2% with GPT 4.1 - a named behavioral improvement |
Trained for Real-World Developer Utility
OpenAI documented that GPT 4.1 was trained with direct input from the developer community, optimizing for tasks like agentic software engineering, extracting insights from large documents, and resolving customer requests. The 1M context window is paired with improved comprehension at long range, not just increased token capacity.
Benchmark Performance
Independent evaluations · Artificial Analysis
Accuracy & Capability Details
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | OpenAI | Anthropic | Anthropic |
| Release Date | April 14, 2025 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | Jun 2024 | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 1.0M Best Context Window | 1M | 1M |
| Pricing (per 1M tokens) | |||
| Input Pricing | $2 Best Input Pricing | $4 | $2 Best Input Pricing |
| Output Pricing | $8 Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | imagetextfile | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 12.7 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | - | - | - |
| Agentic Index | - | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from OpenAI
Other models by OpenAI
Top AI Models
Leading alternatives by intelligence score
