Back to all models
Get all the details on o3-2025-04-16, an AI model from OpenAI. This page covers its token limits, pricing structure, key capabilities such as code_interpretation, web_browsing_via_tool, streaming, available API code samples, and performance strengths.
Key Metrics
Input Limit
200K tokens
Output Limit
100K tokens
Input Cost
$10.00/1M
Output Cost
$40.00/1M
Sample API Code
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(model="o3-2025-04-16", messages=[{"role": "user", "content": "Hello world"}])
print(response.choices[0].message.content)Required Libraries
openai
openai
Benchmarks
| Benchmark | Score | Source | Notes |
|---|---|---|---|
1409 | OpenLLM Leaderboard | Rank 2 overall | |
1190 | OpenLLM Leaderboard | Rank 7 overall | |
1303 | OpenLLM Leaderboard | Rank 2 overall | |
83.3% | Vellum | Best in Reasoning (GPQA Diamond) | |
91.6% | Vellum | Best in High School Math (AIME 2024) | |
69.1% | Vellum | Best in Agentic Coding (SWE Bench) | |
80.71 | LiveBench | Score for 'o3 High' | |
93.33 | LiveBench | Score for 'o3 High' | |
76.71 | LiveBench | Score for 'o3 High' | |
85.00 | LiveBench | Score for 'o3 High' | |
67.02 | LiveBench | Score for 'o3 High' | |
76.00 | LiveBench | Score for 'o3 High' | |
86.17 | LiveBench | Score for 'o3 High' |
Notes
o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following. Use it to think through multi-step problems that involve analysis across text, code, and images.
Capabilities
code interpretation
web browsing via tool
streaming
multimodal input
long context
function calling
structured outputs
advanced reasoning
vision
Supported Data Types
Input Types
text
image
Output Types
text
Strengths & Weaknesses
Exceptional at
math
science
coding
visual reasoning
technical writing
instruction following
multi-step problem solving
multimodal understanding
Good at
general reasoning
well rounded across domains
Additional Information
Latest Update
Apr 16, 2025
Knowledge Cutoff
Jun 1, 2024