LLMcompare

Search

Search for a command to run...

A · closed · GPT-4 / o-series

GPT-4.1

OpenAI

B · closed · GPT-4 / o-series

o3

OpenAI

1:16

overall capability wins (0 ties across 17 scored). o3 leads overall.

  • Reasoning0:9
  • Coding0:5
  • Arena0:1
  • Speed2:0
  • Specs1:1

Verdict

o3 leads coding (widest gap: +35.1 on LiveCode); o3 ranks higher on LMArena (+17 Elo); GPT-4.1 offers ~5.0x larger context (1M vs 200K).

  • o3 leads coding (widest gap: +35.1 on LiveCode)
  • o3 ranks higher on LMArena (+17 Elo)
  • GPT-4.1 offers ~5.0x larger context (1M vs 200K)
  • o3 allows ~3.1x more max output (100K vs 33K)

Cost for 1M in + 250K out

Illustrative chat workload at primary-provider list prices.

GPT-4.1
$4
o3
$4

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 14, 2025

Apr 16, 2025

Knowledge cutoff

2024-06

-

API / provider

OpenAI

OpenAI

Modalities

text, image → text

text, image → text

Specs

Context window

1M

200K

A +800K

Max output

33K

100K

B +67K

Parameters

—

—

—

Pricing

Input $/1M

$2

$2

tie

Output $/1M

$8

$8

tie

Blended $/1M (3∶1)

$3.50

$3.50

tie

Speed

tok/s

95

55

A +40

TTFT (s)

0.35

1.2

A +0.85

Reasoning

MMLU-Pro

80.6

85.3

B +4.7 pts

GPQA Diamond

66.6

82.7

B +16.1 pts

Humanity's Last Exam

4.2

20.1

B +15.9 pts

AIME 2025

46.4

88.9

B +42.5 pts

MATH-500

91.3

99.2

B +7.9 pts

Humanity's Last Exam (with tools)

—

—

—

AA-Omniscience Accuracy

27.8

38.6

B +10.8 pts

AA-LCR v1.1

68.3

74.7

B +6.4 pts

CritPt

—

1.1

—

MMMU-Pro

61.2

70.1

B +8.9 pts

IFBench

43.0

71.4

B +28.4 pts

Chartography

—

—

—

Chartography (With Tools)

—

—

—

Coding

SWE-bench Verified

39.6

58.4

B +18.8 pts

SWE-bench Pro

—

—

—

SWE-bench Multilingual

—

—

—

LiveCodeBench

45.7

80.8

B +35.1 pts

Terminal-Bench 2.1

—

—

—

Aider Polyglot

52.4

81.3

B +28.9 pts

Terminal-Bench 3

—

—

—

BigCodeBench

—

—

—

SciCode

38.1

41.0

B +2.9 pts

CursorBench

—

—

—

SWE-Rebench

30.1

38.1

B +8.0 pts

NL2Repo-Bench

—

—

—

DeepSWE

—

—

—

WebDev Arena

—

—

—

Terminal-Bench 4.0

—

—

—

CursorBench 4.0

—

—

—

FrontierCode v1.1 (Main)

—

—

—

Arena

LMArena Elo

1,415

1,432

B +17 Elo

GPT-4.1

Context
1M / 33K out
Parameters
—
Price
$2 / $8
Speed
95 tok/s · 0.35s TTFT
Modalities
text, image → text
License
Proprietary

Previous-gen OpenAI instruction model with 1M context; still useful for long-document workloads.

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.

Benchmark charts

Winner bars are emphasized. Per-benchmark deltas sit above each chart.

Reasoning & knowledge

MMLU-Pro: B +4.7GPQA: B +16.1HLE: B +15.9AIME: B +42.5MATH-500: B +7.9AA-LCR: B +6.4Omniscience Acc.: B +10.8MMMU-Pro: B +8.9IFBench: B +28.4

Coding

SWE-bench: B +18.8LiveCode: B +35.1Aider: B +28.9SciCode: B +2.9SWE-Rebench: B +8.0

Arena

Arena Elo: B +17

Pick a different pair · Back to catalog