Model comparison
GPT-4.5 vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 37.2 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. GPT-4.5 scores higher in 4 categories and GPT-5.4 nano in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 nano leads 23.7 to 13.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 87.8% for GPT-5.4 nano.
Side by side
| GPT-4.5 | GPT-5.4 nano | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 41.9 |
| Released | 2025-02-27 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.25 |
| Results tracked | 42 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-4.5: 42.2 (#109), GPT-5.4 nano: 43.6 (#84)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| WeirdML | 39.4% | 49.2% |
| LMArena Coding | 1396 | 1405 |
| Aider Polyglot | 44.9% | — |
| SciCode | — | 46.9% |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 1,005 |
Agentic & Tool Use Not comparable
GPT-4.5: 27.9 (#97), GPT-5.4 nano: —
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| Cybench | 17.5% | — |
Reasoning GPT-5.4 nano leads
GPT-4.5: 13.9 (#330), GPT-5.4 nano: 23.7 (#173)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 0.8% | 5.7% |
| ARC-AGI-1 | 10.3% | 51.5% |
| LMArena Hard Prompts | 1403 | 1381 |
| Epoch Capabilities Index | 136.74 | 145.81 |
| ForecastBench | 61.7 | 57.3 |
| SimpleBench | 34.5% | — |
| Kagi LLM Benchmark | — | 39.7% |
| CritPt | — | 9.3% |
| Chess Puzzles | — | 30% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 36.9% |
| LiveBench | 69% | — |
Math GPT-5.4 nano leads
GPT-4.5: 32.6 (#211), GPT-5.4 nano: 40.9 (#88)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 87.8% |
| LMArena Math | 1412 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.4 nano leads
GPT-4.5: 32.5 (#211), GPT-5.4 nano: 41.9 (#103)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 68.7% | 78.5% |
| LMArena Expert | 1394 | 1396 |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | — | 11.7% |
| Confabulations | 13.6% | — |
| Vectara Hallucination Rate | — | 3.1% |
Multimodal Too close to call
GPT-4.5: 37.6 (#71), GPT-5.4 nano: 36.7 (#78)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1195 | 1196 |
| VPCT | 45% | — |
Multilingual GPT-4.5 leads
GPT-4.5: 52.5 (#83), GPT-5.4 nano: 48.6 (#140)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1413 | 1359 |
| LMArena Chinese | 1421 | 1392 |
| LMArena French | 1418 | 1396 |
| LMArena German | 1457 | 1367 |
| LMArena Japanese | 1416 | 1343 |
| LMArena Korean | 1392 | 1320 |
| LMArena Russian | 1419 | 1363 |
| LMArena Spanish | — | 1371 |
Instruction Following Too close to call
GPT-4.5: 72.6 (#134), GPT-5.4 nano: 71.9 (#144)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1404 | 1362 |
| LiveBench Instruction Following | 72.3% | — |
Long Context GPT-5.4 nano leads
GPT-4.5: 40.4 (#155), GPT-5.4 nano: 41.6 (#137)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1406 | 1366 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-4.5 leads
GPT-4.5: 56.9 (#134), GPT-5.4 nano: 55.7 (#142)
| Benchmark | GPT-4.5 | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1417 | 1372 |
| LMArena Creative Writing | 1394 | 1314 |
| LMArena Multi-Turn | 1444 | 1382 |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 37.2 on the Noometry Index.
Is GPT-4.5 or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 42.2 in the Noometry coding category.
How many benchmarks do GPT-4.5 and GPT-5.4 nano share?
24 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5.4 nano has 40.