Compare / head-to-head
GPT-5.4 Provso3
GPT-5.4 Pro leads 2 of 3 shared benchmarks. o3 is 21x cheaper per token. GPT-5.4 Pro has the larger context window (1.1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
2 – 1
GPT-5.4 Pro leads
Cheaper per token
o3
21x cheaper, input + output
Larger context
GPT-5.4 Pro
1.1M tokens
Provider uptime (30d)
94.2446%
OpenAI
GPT-5.4 Pro
OpenAI · released 2026-03-05
textvision
Quality
Benchmark matrix
3 shared · 9 only GPT-5.4 Pro · 4 only o3| Benchmark | GPT-5.4 Pro | o3 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 3 | ||||
| GPQA Diamond | 94.4% ↗ | 81.8% ↗epoch | +12.6 pt | GPT-5.4 Pro |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 82.5% ↗epoch | 33.3% ↗epoch | +49.2 pt | GPT-5.4 Pro |
| SimpleQA Verified | 46.3% ↗epoch | 49.4% ↗epoch | -3.1 pt | o3 |
| Only GPT-5.4 Pro reports · 9 | ||||
| Humanity's Last Exam (no tools) | 42.7% ↗ | not reported | — | — |
| ARC-AGI-1 (Verified) | 94.5% ↗ | not reported | — | — |
| ARC-AGI-2 (Verified) | 83.3% ↗ | not reported | — | — |
| BrowseComp | 89.3% ↗ | not reported | — | — |
| FrontierMath Tier 1-3 | 50% ↗ | not reported | — | — |
| FrontierMath Tier 4 | 38% ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 58.5% ↗epoch | not reported | — | — |
| GDPval (wins or ties) | 82% ↗ | not reported | — | — |
| Humanity's Last Exam (with tools) | 58.7% ↗ | not reported | — | — |
| Only o3 reports · 4 | ||||
| SWE-bench Verified | not reported | 69.1% ↗ | — | — |
| AIME 2025 | not reported | 89.2% ↗matharena ⚠ | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 84.4% ↗epoch | — | — |
| SWE-bench Verified (Epoch AI run) | not reported | 62.3% ↗epoch | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GPT-5.4 Pro minus o3 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | GPT-5.4 Pro | o3 | Edge |
|---|---|---|---|
| Context window | 1.1M tokens | 200K tokens | GPT-5.4 Pro |
| Max output | 128K tokens | 100K tokens | GPT-5.4 Pro |
| Input price / 1M | $30 ↗ | $2 ↗ | o3 |
| Output price / 1M | $180 ↗ | $8 ↗ | o3 |
| Cached input / 1M | — | $0.5 ↗ | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $210 | $10 | o3 |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-03-05 | 2025-04-16 | — |
| Cited benchmark scores | 13 | 7 | — |
Reliability
OpenAI status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
GPT-5.4 Pro vs …
Models sharing the most benchmarksMore matchups
o3 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GPT-5.4 Pro and o3 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.