GPT-5.4 vs GPT-5.4 Pro
| Spec | GPT-5.4 | GPT-5.4 Pro |
|---|---|---|
| Respan Index | 49.5 (#33) | 51.3 (#32) |
| Blended price per 1M | $3.44 | $67.5 |
| Context window | 1.1M | 1.1M |
| Input / output per 1M | $1.25 / $10 | $30 / $180 |
| Max output | 128K | 128K |
| Input | text, vision | text, vision |
| Released | 2026-03-05 | 2026-03-05 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | GPT-5.4 | GPT-5.4 Pro |
|---|---|---|
| Coding | 24.3 GPT-5.4 | 13.5 GPT-5.4 Pro |
| Agents | 62.0 GPT-5.4 | 73.9 GPT-5.4 Pro |
| Reasoning | 61.5 GPT-5.4 | 64.8 GPT-5.4 Pro |
| Knowledge & language | 46.9 GPT-5.4 | 46.4 GPT-5.4 Pro |
| Human preference | 52.6 GPT-5.4 | 57.2* GPT-5.4 Pro |
* Estimated: no published scores in that category. How the Respan Index works
Benchmarks both report
GPT-5.4 Pro leads on 8 of the 10 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | GPT-5.4 | GPT-5.4 Pro |
|---|---|---|
| Chess Puzzles (Epoch AI) | 48% | 59% |
| FrontierMath Tier 4 v2 (Epoch AI) | 49% | 58.5% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 78.6% | 82.5% |
| GPQA Diamond | 92.8% | 94.4% |
| SimpleQA Verified | 45.1% | 46.3% |
| WeirdML (Håvard Tveit Ihle) | 77.7% | 57.4% |
| BrowseComp | 82.7% | 89.3% |
| GDPval (wins or ties) | 83% | 82% |
| Humanity's Last Exam (no tools) | 39.8% | 42.7% |
| Humanity's Last Exam (with tools) | 52.1% | 58.7% |
FAQ
Is GPT-5.4 better than GPT-5.4 Pro?
GPT-5.4 Pro scores 51.3 on the Respan Index (#32), ahead of GPT-5.4 at 49.5 (#33). GPT-5.4 Pro leads on 8 of the 10 benchmarks both report.
Which is cheaper, GPT-5.4 or GPT-5.4 Pro?
GPT-5.4 lists at $1.25 input and $10 output per 1M tokens; GPT-5.4 Pro at $30 and $180. GPT-5.4 is 20x cheaper per token at list price.
Which has the larger context window?
Both accept 1.1M tokens of context.