Menu

logo
Artificial Analysis
HOME
logo

OpenAI: Models Intelligence, Performance & Price

Analysis of OpenAI's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by OpenAI for your use-case. For more details including relating to our methodology, see our FAQs. Models analyzed: o1, o3-mini, o1-preview, o1-mini, GPT-4o (Aug '24), GPT-4o (May '24), GPT-4o (Nov '24), GPT-4o mini, GPT-4.5 (Preview), o3-mini (high), GPT-4 Turbo, GPT-4, and GPT-4o (March 2025).
Link:

OpenAI Model Comparison Summary

Intelligence:o3-mini (high) logo o3-mini (high) and o3-mini logo o3-mini are the highest quality models offered by OpenAI, followed by o1 logo o1, o1-mini logo o1-mini & GPT-4o (March 2025) logo GPT-4o (March 2025).Output Speed (tokens/s):o1-mini logo o1-mini (227 t/s) and GPT-4o (March 2025) logo GPT-4o (March 2025) (226 t/s) are the fastest models offered by OpenAI, followed by o3-mini logo o3-mini, o3-mini (high) logo o3-mini (high) & GPT-4o (Nov '24) logo GPT-4o (Nov '24).Latency (seconds):GPT-4o (March 2025) logo GPT-4o (March 2025) (0.36s) and  GPT-4o mini logo GPT-4o mini (0.37s) are the lowest latency models offered by OpenAI, followed by GPT-4o (Aug '24) logo GPT-4o (Aug '24), GPT-4o (May '24) logo GPT-4o (May '24) & GPT-4o (Nov '24) logo GPT-4o (Nov '24).Blended Price ($/M tokens):GPT-4o mini logo GPT-4o mini ($0.26) and o3-mini logo o3-mini ($1.93) are the cheapest models offered by OpenAI, followed by o1-mini logo o1-mini, o3-mini (high) logo o3-mini (high) & GPT-4o (Aug '24) logo GPT-4o (Aug '24).Context Window Size:o1 logo o1 (200k) and o3-mini logo o3-mini (200k) are the largest context window models offered by OpenAI, followed by o3-mini (high) logo o3-mini (high), o1-preview logo o1-preview & o1-mini logo o1-mini.

Highlights

Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Features
Model Intelligence
Price
Output tokens/s
Latency
End-to-End Response Time
Further
Analysis
OpenAI logo
OpenAI logo
o3-mini (high)
200k
66
$1.93
186.1
39.55
42.24
N/A
OpenAI logo
OpenAI logo
o3-mini
200k
63
$1.93
198.8
13.87
16.39
N/A
OpenAI logo
OpenAI logo
o1
200k
62
$26.25
100.9
29.78
34.74
N/A
OpenAI logo
OpenAI logo
o1-mini
128k
54
$1.93
226.5
10.52
12.72
N/A
OpenAI logo
OpenAI logo
GPT-4o (March 2025)
128k
50
$7.50
226.0
0.36
2.57
N/A
OpenAI logo
OpenAI logo
GPT-4o (Nov '24)
128k
41
$4.38
170.9
0.54
3.47
N/A
OpenAI logo
OpenAI logo
GPT-4o (May '24)
128k
41
$7.50
107.5
0.49
5.14
N/A
OpenAI logo
OpenAI logo
GPT-4o mini
128k
36
$0.26
77.9
0.37
6.79
N/A
OpenAI logo
OpenAI logo
o1-preview
128k
$26.25
128.0
21.61
25.52
N/A
OpenAI logo
OpenAI logo
GPT-4o (Aug '24)
128k
$4.38
120.9
0.49
4.62
N/A
OpenAI logo
OpenAI logo
GPT-4.5 (Preview)
128k
$93.75
11.4
1.48
45.47
N/A
OpenAI logo
OpenAI logo
GPT-4 Turbo
128k
$15.00
47.4
0.66
11.22
N/A
OpenAI logo
OpenAI logo
GPT-4
8k
$37.50
33.9
0.69
15.43
N/A

Key definitions

Context window: Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).
Output Speed: Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Latency (Time to First Token): Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
Price: Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Output Price: Price per token generated by the model (received from the API), represented as USD per million Tokens.
Input Price: Price per token included in the request/message sent to the API, represented as USD per million Tokens.
Time period: Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.