OpenAIGPT-4.1Arena #52Apr 14, 2025

GPT-4.1

GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and GPT-4.5 across coding (54.6% SWE-bench Verified), instruction compliance (87.4% IFEval), and multimodal understanding benchmarks. It is tuned for precise code diffs, agent reliability, and high recall in large document contexts, making it ideal for agents, IDE tooling, and enterprise knowledge retrieval.

Context Window
1.0M
tokens
Max Output
33K
tokens
Released
Apr 14, 2025
Arena Rank
#52
of 305 models
Output Speed
105
tokens/sec
Time to First Token
380ms
TTFT

Capabilities

👁Vision
🧠Reasoning
🔧Tool Calling
Prompt Caching
🖥Computer Use
🎨Image Generation

Supported Parameters

Max Tokens
Output length limit
Response Format
JSON mode / structured output
Seed
Deterministic outputs
structured_outputs
Temperature
Controls randomness
Tool Choice
Control tool usage
Tool Calling
Function calling support
Top P
Nucleus sampling

Pricing Comparison

RouterInput / 1MOutput / 1MCached Input / 1M
Requesty$2.00$8.00$0.50
OpenRouter$2.00$8.00$0.50
Vercel AI$2.00$8.00
Martian$2.00$8.00$0.50

Benchmarks

Artificial Analysis
Intelligence IndexArtificial Analysis
57/100
Coding IndexArtificial Analysis
51/100
Math IndexArtificial Analysis
63/100
MMLU-PROArtificial Analysis
0.748/1
GPQA DiamondArtificial Analysis
0.575/1
MATH-500Artificial Analysis
0.792/1
AIME 2024Artificial Analysis
0.15/1
LiveCodeBenchArtificial Analysis
0.475/1
Aider Polyglot
Aider PolyglotAider Polyglot
20/100

Model IDs

Requestyazure/gpt-4.1@uksouth
OpenRouteropenai/gpt-4.1

Tags

visiontool-callingcaching

Available Regions

EUUS
Compare with another model

Related Models