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
file, image, text
Radar de capacités
39
general
32
coding
49
reasoning
44
scienceest.
60
agents
85
multimodal
Science utilise un proxy de raisonnement lorsque les benchmarks scientifiques dédiés ne sont pas disponibles.
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Code Ranking | 177 | 45.0 | AA |
| General Ranking | 181 | 52.0 | AA |
| Math Reasoning | 188 | 48.0 | AA |
| Multimodal Ranking | 53 | 74.0 | LS |
| Reasoning | 63 | 60.0 | LS |
| Science | 206 | 47.0 | AA |
Scores de benchmarks (LLM Stats)
Biology
GPQA
66.3%Aut.
Code
SWE-Bench Verified
54.6%Aut.
Aider-Polyglot Edit
52.9%Aut.
Aider-Polyglot
51.6%Aut.
Communication
Multi-IF
70.8%Aut.
TAU-bench Retail
68.0%Aut.
TAU-bench Airline
49.4%Aut.
Multi-Challenge
38.3%Aut.
Finance
MMLU
90.2%Aut.
General
IFEval
87.4%Aut.
MMMLU
87.3%Aut.
MMMU
74.8%Aut.
Internal API instruction following (hard)
49.1%Aut.
Language
COLLIE
65.8%Aut.
Long Context
ComplexFuncBench
65.5%Aut.
OpenAI-MRCR: 2 needle 128k
57.2%Aut.
OpenAI-MRCR: 2 needle 1M
46.3%Aut.
Graphwalks parents >128k
25.0%Aut.
Graphwalks BFS >128k
19.0%Aut.
Math
MathVista
72.2%Aut.
AIME 2024
48.1%Aut.
AIME 2025
46.4%Aut.
HMMT 2025
28.9%Aut.
Humanity's Last Exam
5.4%Aut.
Multimodal
CharXiv-D
87.9%Aut.
Video-MME (long, no subtitles)
72.0%Aut.
CharXiv-R
56.7%Aut.
Reasoning
Graphwalks BFS <128k
61.7%Aut.
Graphwalks parents <128k
58.0%Aut.
Indices d'évaluation AA
Math Index34.7
Intelligence Index26.3
Coding Index21.8
Math 5000.9
Mmlu Pro0.8
Gpqa0.7
Lcr0.6
Tau20.5
Livecodebench0.5
Aime0.4
Ifbench0.4
Scicode0.4
Aime 250.3
Terminalbench Hard0.1
Hle0.0
Scores par catégorie LLM Stats
Finance90
Legal90
Healthcare80
Instruction Following80
Language80
Structured Output70
Writing70
Biology70
Chemistry70
General70
Multimodal70
Physics70
Tool Calling60
Vision60
Communication60
Reasoning60
Code50
Frontend Development50
Math50
Spatial Reasoning40
Long Context40
Tarification
Prix d'entrée$2 / 1M tokens
Prix de sortie$8 / 1M tokens
Prix mixte (3:1)$3.5 / 1M tokens
Vitesse
Tokens/sec108.1 tokens/s
Délai du premier token0.55s
Temps de réponse0.55s
Fournisseurs disponibles
(Unités internes LS)| Fournisseur | Prix d'entrée | Prix de sortie |
|---|---|---|
| OpenAI | 2.0M | 8.0M |