vs
0:3
overall capability wins (0 ties across 3 scored). o3-pro leads overall.
- Reasoning0:1
- Pricing3:0
- Speed2:0
- Specs0:2
Verdict
o3-pro edges reasoning & knowledge; o3-pro offers ~1.6x larger context (200K vs 128K); o3-pro allows ~24.4x more max output (100K vs 4K).
- o3-pro edges reasoning & knowledge
- o3-pro offers ~1.6x larger context (200K vs 128K)
- o3-pro allows ~24.4x more max output (100K vs 4K)
- GPT-4o (May 2024) is ~4.7x cheaper on a blended token basis than o3-pro
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- GPT-4o (May 2024)
- $8.75
- o3-pro
- $40
GPT-4o (May 2024) is about 4.6x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
OpenAI
OpenAI
Family
GPT-4 / o-series
GPT-4 / o-series
License
Proprietary
Proprietary
Open weights
No
No
Release
May 13, 2024
Jun 10, 2025
Knowledge cutoff
2023-10
-
API / provider
OpenAI
OpenAI
Modalities
text, image → text
text, image → text
Specs
Context window
128K
200K
Max output
4K
100K
Parameters
—
—
Pricing
Input $/1M
$5
$20
Output $/1M
$15
$80
Blended $/1M (3∶1)
$7.50
$35
Speed
tok/s
100
40
TTFT (s)
0.35
2.5
Reasoning
MMLU-Pro
74.0
—
GPQA Diamond
52.6
84.5
Humanity's Last Exam
1.8
—
AIME 2025
—
—
MATH-500
79.1
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
—
AA-LCR v1.1
—
—
CritPt
—
—
MMMU-Pro
—
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
—
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
33.4
—
Terminal-Bench 2.1
—
—
Aider Polyglot
—
84.9
Terminal-Bench 3
—
—
BigCodeBench
51.1
—
SciCode
30.9
—
CursorBench
—
—
SWE-Rebench
—
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,346
—
| Metric | GPT-4o (May 2024) | o3-pro | Delta |
|---|---|---|---|
| Identity | |||
| Organization | OpenAI | OpenAI | — |
| Family | GPT-4 / o-series | GPT-4 / o-series | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | May 13, 2024 | Jun 10, 2025 | — |
| Knowledge cutoff | 2023-10 | - | — |
| API / provider | OpenAI | OpenAI | — |
| Modalities | text, image → text | text, image → text | — |
| Specs | |||
| Context window | 128K | 200K | B +72K |
| Max output | 4K | 100K | B +96K |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $5 | $20 | A +$15 |
| Output $/1M | $15 | $80 | A +$65 |
| Blended $/1M (3∶1) | $7.50 | $35 | A +$27.50 |
| Speed | |||
| tok/s | 100 | 40 | A +60 |
| TTFT (s) | 0.35 | 2.5 | A +2.15 |
| Reasoning | |||
| MMLU-Pro | 74.0 | — | — |
| GPQA Diamond | 52.6 | 84.5 | B +31.9 pts |
| Humanity's Last Exam | 1.8 | — | — |
| AIME 2025 | — | — | — |
| MATH-500 | 79.1 | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | — | — |
| AA-LCR v1.1 | — | — | — |
| CritPt | — | — | — |
| MMMU-Pro | — | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | — | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 33.4 | — | — |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | — | 84.9 | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | 51.1 | — | — |
| SciCode | 30.9 | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | — | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,346 | — | — |
GPT-4o (May 2024)
- Context
- 128K / 4K out
- Parameters
- —
- Price
- $5 / $15
- Speed
- 100 tok/s · 0.35s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Original gpt-4o-2024-05-13 launch snapshot — higher list rates than later GPT-4o aliases; kept for pinned-version reproducibility.
o3-pro
- Context
- 200K / 100K out
- Parameters
- —
- Price
- $20 / $80
- Speed
- 40 tok/s · 2.5s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Maximum o3 reasoning tier for hardest STEM and multi-step problems at premium $20/$80 rates.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.