Passer au contenu principal

GPT-4.1

OpenAIGPTProprietary

Description

GPT-4.1 is OpenAI's latest and most advanced flagship model, significantly improving upon GPT-4 Turbo in performance across benchmarks, speed, and cost-effectiveness.

Date de sortie
2025-04-14
Paramètres
Longueur du contexte
1.0M
Modalités
image, pdf, text

Radar de capacités

36
general
44
coding
49
reasoning
44
science
60
agents
85
multimodal

Classements

Domaine#RangScoreSource
Classement codage237
49.0
AA
Classement général257
48.0
AA
Classement multimodal55
44.0
LS
Science302
45.0
AA

Scores de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)66.3%Aut.

Code

SWE-Bench Verified54.6%Aut.
Aider-Polyglot Edit52.9%Aut.
Aider-Polyglot51.6%Aut.

Communication

Multi-IF70.8%Aut.
TAU-bench Retail68.0%Aut.
TAU-bench Airline49.4%Aut.
Multi-Challenge38.3%Aut.

Finance

MMLU90.2%Aut.

General

IFEvalGoogle Research (2023)87.4%Aut.
MMMLU87.3%Aut.
MMMU74.8%Aut.
Internal API instruction following (hard)49.1%Aut.

Language

COLLIE65.8%Aut.

Long Context

ComplexFuncBench65.5%Aut.
OpenAI-MRCR: 2 needle 128k57.2%Aut.
OpenAI-MRCR: 2 needle 1M46.3%Aut.
Graphwalks parents >128k25.0%Aut.
Graphwalks BFS >128k19.0%Aut.

Math

MathVista72.2%Aut.
AIME 202448.1%Aut.
AIME 202546.4%Aut.
HMMT 202528.9%Aut.
Humanity's Last Exam5.4%Aut.

Multimodal

CharXiv-D87.9%Aut.
Video-MME (long, no subtitles)72.0%Aut.
CharXiv-R56.7%Aut.

Reasoning

Graphwalks BFS <128k61.7%Aut.
Graphwalks parents <128k58.0%Aut.

Indices d'évaluation AA

(Artificial Analysis)
Math Index(Artificial Analysis)
34.7
Intelligence Index(Artificial Analysis)
19.6
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.8
Gpqa(NYU + Cohere + Anthropic (2023))
0.7
Lcr(Artificial Analysis)
0.6
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.5
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.5
Aime(MAA (Mathematical Association of America))
0.4
Ifbench(Google Research (2023))
0.4
Scicode(UIUC + Argonne National Lab (2024))
0.4
Aime 25(MAA (Mathematical Association of America))
0.3
Terminalbench Hard(Stanford × Laude Institute (2026))
0.1
Hle(Center for AI Safety + Scale AI (2025))
0.0

Scores par catégorie LLM Stats

(LLM Stats (zeroeval))
Legal
90
Finance
90
Instruction Following
80
Language
80
Healthcare
80
Multimodal
70
Physics
70
Structured Output
70
Biology
70
Chemistry
70
Writing
70
Reasoning
60
General
60
Communication
60
Tool Calling
60
Vision
60
Math
50
Frontend Development
50
Code
50
Long Context
40
Spatial Reasoning
40

Tarification

Prix d'entrée$2 / 1M tokens
Prix de sortie$8 / 1M tokens
Prix mixte (3:1)$3.5 / 1M tokens
Prix de lecture cache$0.5 / 1M tokens

Vitesse

Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s

Classement des Prix par Fournisseur

Classement des Prix par Fournisseur

22 fournisseurs

Moins cher: OpenAIPlus cher: Cortecs
FournisseurEntréeSortie
1OpenAIMoins cher
$0
$0.00001
2Poe
$1.8
$7.2
3302.AI
$2
$8
4NanoGPT
$2
$8
5Abacus
$2
$8
6OpenRouter
$2
$8
7Kilo Gateway
$2
$8
8SAP AI Core
$2
$8
9GitHub Copilot
$2
$8
10Helicone
$2
$8
11Azure Cognitive Services
$2
$8
12Vercel AI Gateway
$2
$8
13LLM Gateway
$2
$8
14Azure
$2
$8
15FastRouter
$2
$8
16NEAR AI Cloud
$2
$8
17OrcaRouter
$2
$8
18Merge Gateway
$2
$8
19Pioneer
$2
$8
20Ofox
$2
$8
21Impossibl
$2
$8
22Cortecs
$2.192
$8.769

Comparer les prix entre différents fournisseurs API pour ce modèle.

Sources externes