GPT-4
OpenAIGPTProprietary
Description
GPT-4 is a large multimodal model capable of processing both image and text inputs and generating human-like text outputs. It demonstrates human-level performance on various professional and academic benchmarks.
Release Date
2023-03-14
Parameters
—
Context Length
8K
Modalities
image, text
Capability Radar
13
general
13
coding
70
reasoning
34
scienceest.
0
agents
80
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 323 | 22.0 | AA |
| General Ranking | 466 | 15.0 | AA |
| Reasoning | 3 | 95.0 | LS |
Benchmark Scores (LLM Stats)
Biology
GPQA
35.7%SR
Code
HumanEval
67.0%SR
Finance
MMLU
86.4%SR
General
AI2 Reasoning Challenge (ARC)
96.3%SR
LSAT
88.0%SR
Language
Winogrande
87.5%SR
Legal
Uniform Bar Exam
90.0%SR
Math
SAT Math
89.0%SR
DROP
80.9%SR
MGSM
74.5%SR
MATH
42.0%SR
Reasoning
HellaSwag
95.3%SR
AA Evaluation Indices
Coding Index13.1
Intelligence Index12.8
LLM Stats Category Scores
Finance90
Healthcare90
Language90
Legal90
Code70
Math70
Reasoning70
General60
Biology40
Chemistry40
Physics40
Pricing
Input Price$30 / 1M tokens
Output Price$60 / 1M tokens
Blended Price (3:1)$37.5 / 1M tokens
Speed
Tokens/sec24.1 tokens/s
Time to First Token1.00s
Time to Answer1.00s
Available Providers
(LS internal units)| Provider | Input Price | Output Price |
|---|---|---|
| OpenAI | 30.0M | 60.0M |
| Azure | 30.0M | 60.0M |