vs
8:0
overall capability wins (2 ties across 10 scored). o3 leads overall.
- Reasoning4:0
- Coding3:0
- Arena1:0
- Pricing0:3
- Speed0:2
Verdict
o3 leads coding (widest gap: +27.5 on Aider); o3 ranks higher on LMArena (+84 Elo); o3-mini is ~1.8x cheaper on a blended token basis than o3.
- o3 leads coding (widest gap: +27.5 on Aider)
- o3 ranks higher on LMArena (+84 Elo)
- o3-mini is ~1.8x cheaper on a blended token basis than o3
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- o3
- $4
- o3-mini
- $2.20
o3-mini is about 1.8x 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
Apr 16, 2025
Jan 31, 2025
Knowledge cutoff
-
-
API / provider
OpenAI
OpenAI
Modalities
text, image → text
text → text
Specs
Context window
200K
200K
Max output
100K
100K
Parameters
—
—
Pricing
Input $/1M
$2
$1.10
Output $/1M
$8
$4.40
Blended $/1M (3∶1)
$3.50
$1.93
Speed
tok/s
55
100
TTFT (s)
1.2
0.6
Reasoning
MMLU-Pro
85.3
79.1
GPQA Diamond
82.7
74.8
Humanity's Last Exam
20.1
7.9
AIME 2025
88.9
—
MATH-500
99.2
97.3
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
38.6
—
AA-LCR v1.1
74.7
—
CritPt
1.1
—
MMMU-Pro
70.1
—
IFBench
71.4
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
58.4
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
80.8
71.7
Terminal-Bench 2.1
—
—
Aider Polyglot
81.3
53.8
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
41.0
39.9
CursorBench
—
—
SWE-Rebench
38.1
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,432
1,348
| Metric | o3 | o3-mini | Delta |
|---|---|---|---|
| Identity | |||
| Organization | OpenAI | OpenAI | — |
| Family | GPT-4 / o-series | GPT-4 / o-series | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | Apr 16, 2025 | Jan 31, 2025 | — |
| Knowledge cutoff | - | - | — |
| API / provider | OpenAI | OpenAI | — |
| Modalities | text, image → text | text → text | — |
| Specs | |||
| Context window | 200K | 200K | tie |
| Max output | 100K | 100K | tie |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $2 | $1.10 | B +$0.90 |
| Output $/1M | $8 | $4.40 | B +$3.60 |
| Blended $/1M (3∶1) | $3.50 | $1.93 | B +$1.57 |
| Speed | |||
| tok/s | 55 | 100 | B +45 |
| TTFT (s) | 1.2 | 0.6 | B +0.60 |
| Reasoning | |||
| MMLU-Pro | 85.3 | 79.1 | A +6.2 pts |
| GPQA Diamond | 82.7 | 74.8 | A +7.9 pts |
| Humanity's Last Exam | 20.1 | 7.9 | A +12.2 pts |
| AIME 2025 | 88.9 | — | — |
| MATH-500 | 99.2 | 97.3 | A +1.9 pts |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 38.6 | — | — |
| AA-LCR v1.1 | 74.7 | — | — |
| CritPt | 1.1 | — | — |
| MMMU-Pro | 70.1 | — | — |
| IFBench | 71.4 | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | 58.4 | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 80.8 | 71.7 | A +9.1 pts |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | 81.3 | 53.8 | A +27.5 pts |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 41.0 | 39.9 | A +1.1 pts |
| CursorBench | — | — | — |
| SWE-Rebench | 38.1 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,432 | 1,348 | A +84 Elo |
o3
- Context
- 200K / 100K out
- Parameters
- —
- Price
- $2 / $8
- Speed
- 55 tok/s · 1.2s TTFT
- Modalities
- text, image → text
- License
- Proprietary
OpenAI reasoning model from the o-series; strong on STEM and agentic coding with deliberate chain-of-thought.
o3-mini
- Context
- 200K / 100K out
- Parameters
- —
- Price
- $1.10 / $4.40
- Speed
- 100 tok/s · 0.6s TTFT
- Modalities
- text → text
- License
- Proprietary
Earlier o-series mini - still useful for math-heavy workloads on a budget.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.