vs
3:1
overall capability wins (0 ties across 4 scored). GPT-4.1 leads overall.
- Coding1:0
- Arena0:1
- Pricing3:0
- Speed2:0
- Specs2:0
Verdict
GPT-4.1 leads coding (widest gap: +7.5 on Aider); GPT-4.5 ranks higher on LMArena (+30 Elo); GPT-4.1 offers ~7.8x larger context (1M vs 128K).
- GPT-4.1 leads coding (widest gap: +7.5 on Aider)
- GPT-4.5 ranks higher on LMArena (+30 Elo)
- GPT-4.1 offers ~7.8x larger context (1M vs 128K)
- GPT-4.1 allows ~2.0x more max output (33K vs 16K)
- GPT-4.1 is ~26.8x cheaper on a blended token basis than GPT-4.5
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- GPT-4.1
- $4
- GPT-4.5
- $112.50
GPT-4.1 is about 28.1x 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 14, 2025
Feb 27, 2025
Knowledge cutoff
2024-06
-
API / provider
OpenAI
OpenAI
Modalities
text, image → text
text, image → text
Specs
Context window
1M
128K
Max output
33K
16K
Parameters
—
—
Pricing
Input $/1M
$2
$75
Output $/1M
$8
$150
Blended $/1M (3∶1)
$3.50
$93.75
Speed
tok/s
95
45
TTFT (s)
0.35
0.7
Reasoning
MMLU-Pro
80.6
—
GPQA Diamond
66.6
—
Humanity's Last Exam
4.2
—
AIME 2025
46.4
—
MATH-500
91.3
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
27.8
—
AA-LCR v1.1
68.3
—
CritPt
—
—
MMMU-Pro
61.2
—
IFBench
43.0
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
39.6
—
SWE-bench Pro
—
—
SWE-bench Multilingual
—
—
LiveCodeBench
45.7
—
Terminal-Bench 2.1
—
—
Aider Polyglot
52.4
44.9
Terminal-Bench 3
—
—
BigCodeBench
—
—
SciCode
38.1
—
CursorBench
—
—
SWE-Rebench
30.1
—
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,415
1,445
| Metric | GPT-4.1 | GPT-4.5 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | OpenAI | OpenAI | — |
| Family | GPT-4 / o-series | GPT-4 / o-series | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | Apr 14, 2025 | Feb 27, 2025 | — |
| Knowledge cutoff | 2024-06 | - | — |
| API / provider | OpenAI | OpenAI | — |
| Modalities | text, image → text | text, image → text | — |
| Specs | |||
| Context window | 1M | 128K | A +872K |
| Max output | 33K | 16K | A +16K |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $2 | $75 | A +$73 |
| Output $/1M | $8 | $150 | A +$142 |
| Blended $/1M (3∶1) | $3.50 | $93.75 | A +$90.25 |
| Speed | |||
| tok/s | 95 | 45 | A +50 |
| TTFT (s) | 0.35 | 0.7 | A +0.35 |
| Reasoning | |||
| MMLU-Pro | 80.6 | — | — |
| GPQA Diamond | 66.6 | — | — |
| Humanity's Last Exam | 4.2 | — | — |
| AIME 2025 | 46.4 | — | — |
| MATH-500 | 91.3 | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 27.8 | — | — |
| AA-LCR v1.1 | 68.3 | — | — |
| CritPt | — | — | — |
| MMMU-Pro | 61.2 | — | — |
| IFBench | 43.0 | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | 39.6 | — | — |
| SWE-bench Pro | — | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 45.7 | — | — |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | 52.4 | 44.9 | A +7.5 pts |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | — | — | — |
| SciCode | 38.1 | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | 30.1 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,415 | 1,445 | B +30 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.
GPT-4.5
- Context
- 128K / 16K out
- Parameters
- —
- Price
- $75 / $150
- Speed
- 45 tok/s · 0.7s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Research preview bridging GPT-4o and GPT-5 — stronger world knowledge and writing, at premium API rates.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.