NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/GPT-4.1
OpenAI logo

GPT-4.1

by OpenAI · Released Apr 2025

Multimodal1M Context
37.9
avg score
Rank #205
Compare
Better than 34% of all models
Context
1.0M tokens (~524 books)
Input $/1M
$2.00
Output $/1M
$8.00
Type
multimodal
License
Proprietary
Benchmarks
27 tested
Data as of
About

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...

Tested on 27 benchmarks · BenchGecko score 37.9. Top scores: HELM — WildBench (85.4%), HELM — IFEval (83.8%), MATH level 5 (83.0%).

Looking for similar performance at lower cost?
GPT-5 Nano scores 37.7 (99% as good) at $0.05/1M input · 98% cheaper
Capabilities
coding
44.3
#122 globally
reasoning
25.9
#134 globally
math
30.0
#191 globally
knowledge
44.5
#168 globally
general
38.6
#75 globally
language
83.8
#42 globally
Benchmark Scores
Compare All
Tested on 27 benchmarks · Ranked across 6 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
Aider polyglot

Multi-language code editing from Aider. Tests editing ability across Python, JavaScript, TypeScript, Java, C++, Go, Rust, and more.

52.4·
SWE-Bench verified

Real-world software engineering tasks from GitHub issues. Models must diagnose bugs and write patches that pass test suites. Human-verified subset of SWE-bench.

48.5·
CadEval

Computer-aided design evaluation. Tests understanding of CAD concepts, 3D modeling, and engineering design principles.

42.0·
reasoningCompare reasoning →
HELM — WildBench

Stanford HELM WildBench evaluation. Tests reasoning on challenging real-world tasks.

85.4·
SimpleBench

Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.

12.4·
ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

5.5·
mathCompare math →
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

83.0·
HELM — Omni-MATH

Stanford HELM evaluation of mathematical reasoning across diverse problem types.

47.1·
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

38.3·
Quick compare:nousresearch logovs Hermes 2 Pro - Llama-3 8BAnthropic logovs Claude 2OpenAI logovs GPT-5 Nano
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · OpenAI GPT-4.1
Apr 2025
37.9
$2.00/M in1.0Mctx27 benchmarks
Apr 2025
$1.00/M in(-1)1.0Mctx
Apr 2025
35.2-2.7
$0.40/M in(-0.60)1.0Mctx22 benchmarks
Apr 2025
$0.20/M in(-0.20)1.0Mctx
Apr 2025
22.4-12.8
$0.10/M in(-0.10)1.0Mctx17 benchmarks
Apr 2025
$0.05/M in(-0.05)1.0Mctx
See the full GPT-4.1 family →
Similar Models
nousresearch logo
Hermes 2 Pro - Llama-3 8B
nousresearch
37.8$0.14/1M
Anthropic logo
Claude 2
Anthropic
37.7n/a
OpenAI logo
GPT-5 Nano
OpenAI
37.7$0.05/1M
Links
Info
OpenAIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@OpenAI
BenchGecko API
gpt-4-1
Specifications
  • Typemultimodal
  • Context1.0M tokens (~524 books)
  • ReleasedApr 2025
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.012
Available On
OpenAI logoOpenAI$2.00
Categories
codingreasoningmathknowledgegenerallanguage
Powers Agents
SWE-agent66.6%View all agents →
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
nousresearch logoHermes 2 Pro - Llama-3 8B
37.8
Anthropic logoClaude 2
37.7
OpenAI logoGPT-5 Nano
37.7
z-ai logoGLM 4 32B
37.5
GPT-4.1 is a proprietary multimodal AI model by OpenAI, released in April 2025. It has an average benchmark score of 37.9. Context window: 1M tokens.

Related Models

Hermes 2 Pro - Llama-3 8B · nousresearchClaude 2 · AnthropicGPT-5 Nano · OpenAIGLM 4 32B · z-aiGemini 2.5 Flash · Google DeepMind

Benchmarks

HELM — WildBenchHELM — IFEvalMATH level 5HELM — MMLU-ProGeoBench

Related Pages

OpenAI · ProviderOpenAI · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • GPT-4.1 by OpenAI. BenchGecko score 37.9, rank 205 of 312 scored models (normalized average of public benchmark scores).
  • List price $2.00 input · $8.00 output per 1M tokens (as of 2026-10-05).
  • Sold by 3 providers (as of 2026-10-05): Azure $2.00 in / $8.00 out · OpenAI $2.00 in / $8.00 out · Azure $2.20 in / $8.80 out. Every provider

How to cite · data as of 2026-10-05

GPT-4.1 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/gpt-4-1

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP