GPT-5.4 (2026-03-05) vs GPT-5.5
MirrorCode?
GPT-5.4 (2026-03-05)at high15.6
GPT-5.5at high10
MultiLangBench?lab
GPT-5.4 (2026-03-05)—
GPT-5.525.8
| Benchmark | GPT-5.4 (2026-03-05) | GPT-5.5 |
|---|---|---|
| Terminal-Bench 4.0 | — | 5.1 |
| MirrorCode | 15.6 · high | 10 · high |
| DeepSWE 1.1 | — | 54 |
| DeepSWE 1 | — | 48 |
| SciCode | — | 54.5 |
| Terminal-Bench Hard (version unstated) | — | 57.6 |
| MultiLangBenchlab | — | 25.8 |
Specs
| Spec | GPT-5.4 (2026-03-05) | GPT-5.5 |
|---|---|---|
| Input, per 1M | !$2.50 | $5 |
| Output, per 1M | !$15 | $30 |
| Cost per task? | — | $0.77 |
| Speed, tok/s | — | 82 |
| Context | 1.05M | 1.05M |
| Weights | Not verified | Closed |
| Released | Mar 2026 | Apr 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ? | Partial |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | ? | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
GPT-5.4 (2026-03-05) → GPT-5.5
Prompts with >272K input tokens›
Above 272K input tokens, input costs 2× and output 1.5× for the full session in Standard, Batch and Flex.
GPT-5.5 → GPT-5.4 (2026-03-05)
Nothing known.