Model comparison
GPT-5.4 mini vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 45.0 on the Noometry Index. GPT-5.4 mini costs 6.7× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 45 shared benchmarks.
Summary
- They share 45 benchmarks with published results for both. GPT-5.4 mini scores higher in 0 categories and GPT-5.5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 30.4.
- The biggest single-benchmark swing is ARC-AGI-2: 18.9% for GPT-5.4 mini and 85% for GPT-5.5.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5.4 mini | GPT-5.5 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 45.0 | 63.4 |
| Released | 2026-03-17 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.75 | $5 |
| Output $ / M tokens | $4.50 | $30 |
| Results tracked | 46 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
GPT-5.4 mini: 45.2 (#72), GPT-5.5: 58.2 (#17)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| FrontierCode | 27% | 43% |
| LMArena WebDev | 1397 | 1513 |
| SciCode | 49.9% | 56.1% |
| WeirdML | 60.3% | 84.9% |
| LMArena Coding | 1438 | 1494 |
| ALE-Bench | 1,189 | 1,943 |
| SWE-bench Verified | — | 80.6% |
| DeepSWE | — | 67% |
| GSO | — | 40.2% |
| MirrorCode | — | 10% |
Agentic & Tool Use GPT-5.5 leads
GPT-5.4 mini: 29.9 (#81), GPT-5.5: 50.7 (#6)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| DeepResearch Bench | 36.3% | 54% |
| Terminal-Bench | — | 84.7% |
| APEX-Agents | — | 55.1% |
| OSWorld 2.0 | — | 13% |
| Remote Labor Index | — | 6.3% |
| τ²-bench Banking | — | 44.6% |
| PostTrainBench | — | 27.2% |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
| LMArena Search | — | 1242 |
| Vending-Bench 2 | — | 7,524 |
Reasoning GPT-5.5 leads
GPT-5.4 mini: 30.4 (#85), GPT-5.5: 72.8 (#11)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 18.9% | 85% |
| Kagi LLM Benchmark | 37.9% | 88.8% |
| NYT Connections (extended) | 61.8% | 96.2% |
| ARC-AGI-1 | 63.7% | 95% |
| CritPt | 10% | 27.1% |
| Chess Puzzles | 24% | 54% |
| LMArena Hard Prompts | 1424 | 1489 |
| Mystery Game Puzzles | 11% | 56% |
| DTBench | 80% | 96% |
| LMCA | 40.8% | 54.3% |
| Epoch Capabilities Index | 148.84 | 159.1 |
| ForecastBench | 57 | 60.6 |
| SimpleBench | — | 69% |
| Thematic Generalization | 61.7% | — |
| EBR-Bench | — | 34.3% |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
Math GPT-5.5 leads
GPT-5.4 mini: 45.5 (#75), GPT-5.5: 81.7 (#11)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 85.3% |
| FrontierMath Tier 4 | 9.8% | 72.5% |
| OTIS Mock AIME 2024-2025 | 88.9% | 100% |
| ProofBench | 21% | 50% |
| LMArena Math | 1419 | 1486 |
| FrontierMath (Feb 2025 set) | 28.3% | 51.7% |
| FrontierMath Tier 4 (v1) | 2.1% | 35.4% |
| MathArena Final-Answer Competitions | — | 94.3% |
| FrontierMath Erdős | — | 0% |
Knowledge GPT-5.5 leads
GPT-5.4 mini: 51.5 (#67), GPT-5.5: 64.4 (#17)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 86.9% | 94% |
| SimpleQA Verified | 29.4% | 63% |
| Vectara Hallucination Rate | 5.5% | 9.3% |
| LMArena Expert | 1435 | 1508 |
Multimodal GPT-5.5 leads
GPT-5.4 mini: 39.7 (#56), GPT-5.5: 46.9 (#12)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1245 | 1297 |
| Blueprint-Bench 2 | — | 36.2% |
| Furniture Assembly | — | 44.2% |
| LMArena Document | — | 1486 |
Multilingual GPT-5.5 leads
GPT-5.4 mini: 51.9 (#96), GPT-5.5: 56.4 (#20)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1405 | 1467 |
| LMArena Chinese | 1446 | 1533 |
| LMArena French | 1440 | 1486 |
| LMArena German | 1409 | 1480 |
| LMArena Japanese | 1374 | 1498 |
| LMArena Korean | 1368 | 1460 |
| LMArena Russian | 1417 | 1473 |
| LMArena Spanish | 1405 | 1468 |
Instruction Following GPT-5.5 leads
GPT-5.4 mini: 74.1 (#102), GPT-5.5: 77.5 (#18)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1479 |
Long Context GPT-5.5 leads
GPT-5.4 mini: 43.0 (#112), GPT-5.5: 48.3 (#12)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1407 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference GPT-5.5 leads
GPT-5.4 mini: 64.0 (#58), GPT-5.5: 72.7 (#13)
| Benchmark | GPT-5.4 mini | GPT-5.5 |
|---|---|---|
| LMArena Text | 1412 | 1472 |
| LMArena Creative Writing | 1370 | 1455 |
| EQ-Bench Creative Writing | 1665 | 1844 |
| LMArena Multi-Turn | 1429 | 1476 |
| EQ-Bench 4 | — | 1315 |
Frequently asked questions
Is GPT-5.4 mini better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 45.0 on the Noometry Index. GPT-5.4 mini costs 6.7× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or GPT-5.5?
GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; GPT-5.5 lists at $5 and $30.
Is GPT-5.4 mini or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 45.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5.4 mini and GPT-5.5 share?
45 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and GPT-5.5 has 71.