OpenAI

GPT-4

Multimodal
Zero-eval
#1AI2 Reasoning Challenge (ARC)
#1Uniform Bar Exam
#1SAT Math
+3 more

by OpenAI

+
+
+
+
About

GPT-4 was created as a large multimodal model capable of accepting image and text inputs while producing text outputs. Developed to exhibit human-level performance on various professional and academic benchmarks, it marked a significant advancement in reliability, creativity, and handling of nuanced instructions compared to its predecessors.

+
+
+
+
Pricing Range
Input (per 1M)$30.00 -$30.00
Output (per 1M)$60.00 -$60.00
Providers2
+
+
+
+
Timeline
AnnouncedJun 13, 2023
ReleasedJun 13, 2023
Knowledge CutoffDec 31, 2022
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown

Overall Performance

12 benchmarks
Average Score
77.7%
Best Score
96.3%
High Performers (80%+)
8

Performance Metrics

Max Context Window
65.5K
Avg Throughput
102.0 tok/s
Avg Latency
0ms
+
+
+
+
All Benchmark Results for GPT-4
Complete list of benchmark scores with detailed information
AI2 Reasoning Challenge (ARC)
text
0.96
96.3%
Self-reported
HellaSwag
text
0.95
95.3%
Self-reported
Uniform Bar Exam
text
0.90
90.0%
Self-reported
SAT Math
text
0.89
89.0%
Self-reported
LSAT
text
0.88
88.0%
Self-reported
Winogrande
text
0.88
87.5%
Self-reported
MMLU
text
0.86
86.4%
Self-reported
DROP
text
0.81
80.9%
Self-reported
MGSM
text
0.74
74.5%
Self-reported
HumanEval
text
0.67
67.0%
Self-reported
Showing 1 to 10 of 12 benchmarks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+