Model comparison
GPT-4 Turbo vs GPT-5.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 30.3 on the Noometry Index.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and GPT-5.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.5 to 8.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 100% for GPT-5.5.
- GPT-5.5 is cheaper at $5 / $30 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- GPT-5.5 accepts more context: 1.05M tokens versus 128K.
Side by side
| GPT-4 Turbo | GPT-5.5 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 30.3 | 63.4 |
| Released | 2023-11-06 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 1.05M |
| Max output | 4K | 128K |
| Input $ / M tokens | $10 | $5 |
| Output $ / M tokens | $30 | $30 |
| Results tracked | 36 | 71 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
GPT-4 Turbo: 33.4 (#247), GPT-5.5: 58.3 (#13)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| WeirdML | 18% | 84.9% |
| LMArena Coding | 1268 | 1494 |
| SWE-bench Verified | — | 80.6% |
| DeepSWE | — | 67% |
| FrontierCode | — | 43% |
| LMArena WebDev | — | 1512 |
| SciCode | — | 56.1% |
| GSO | — | 40.2% |
| BigCodeBench Instruct | 48.2% | — |
| MirrorCode | — | 10% |
| BigCodeBench Complete | 58.2% | — |
| ALE-Bench | — | 1,943 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, GPT-5.5: 50.7 (#6)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| Terminal-Bench | — | 84.7% |
| APEX-Agents | — | 55.1% |
| OSWorld 2.0 | — | 13% |
| Remote Labor Index | — | 6.3% |
| τ²-bench Banking | — | 44.6% |
| DeepResearch Bench | — | 54% |
| PostTrainBench | — | 27.2% |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
| LMArena Search | — | 1242 |
| METR Time Horizons | 36.7% | — |
| Vending-Bench 2 | — | 7,524 |
Reasoning GPT-5.5 leads
GPT-4 Turbo: 14.9 (#311), GPT-5.5: 72.7 (#8)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| SimpleBench | 25.1% | 69% |
| Chess Puzzles | 6% | 54% |
| LMArena Hard Prompts | 1251 | 1489 |
| DTBench | 61.6% | 96% |
| LMCA | 9.8% | 54.3% |
| Epoch Capabilities Index | 127.25 | 159.25 |
| ForecastBench | 59.4 | 60.6 |
| ARC-AGI-2 | — | 85% |
| Kagi LLM Benchmark | — | 88.8% |
| NYT Connections (extended) | — | 96.2% |
| ARC-AGI-1 | — | 95% |
| CritPt | — | 27.1% |
| EBR-Bench | — | 34.3% |
| Mystery Game Puzzles | — | 56% |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
Math GPT-5.5 leads
GPT-4 Turbo: 8.6 (#316), GPT-5.5: 81.5 (#9)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 85.3% |
| OTIS Mock AIME 2024-2025 | 6.7% | 100% |
| LMArena Math | 1272 | 1486 |
| FrontierMath Tier 4 | — | 72.5% |
| MathArena Final-Answer Competitions | — | 94.3% |
| ProofBench | — | 50% |
| MATH Level 5 | 46.7% | — |
| FrontierMath (Feb 2025 set) | — | 51.7% |
| FrontierMath Erdős | — | 0% |
| FrontierMath Tier 4 (v1) | — | 35.4% |
Knowledge GPT-5.5 leads
GPT-4 Turbo: 24.3 (#263), GPT-5.5: 64.4 (#14)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 46.6% | 94% |
| LMArena Expert | 1223 | 1508 |
| SimpleQA Verified | — | 63% |
| Confabulations | 28.4% | — |
| Vectara Hallucination Rate | — | 9.3% |
| MMLU | 81.3% | — |
Multimodal GPT-5.5 leads
GPT-4 Turbo: 30.0 (#104), GPT-5.5: 47.1 (#11)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1090 | 1297 |
| Blueprint-Bench 2 | — | 36.2% |
| Furniture Assembly | — | 44.2% |
| LMArena Document | — | 1486 |
Multilingual GPT-5.5 leads
GPT-4 Turbo: 40.7 (#211), GPT-5.5: 56.5 (#19)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1245 | 1467 |
| LMArena Chinese | 1242 | 1535 |
| LMArena French | 1275 | 1486 |
| LMArena German | 1259 | 1478 |
| LMArena Japanese | 1194 | 1498 |
| LMArena Korean | 1187 | 1461 |
| LMArena Russian | 1259 | 1473 |
| LMArena Spanish | 1260 | 1469 |
Instruction Following GPT-5.5 leads
GPT-4 Turbo: 66.0 (#211), GPT-5.5: 77.7 (#18)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1250 | 1479 |
Long Context GPT-5.5 leads
GPT-4 Turbo: 38.0 (#200), GPT-5.5: 48.3 (#12)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1254 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference GPT-5.5 leads
GPT-4 Turbo: 47.9 (#201), GPT-5.5: 72.8 (#13)
| Benchmark | GPT-4 Turbo | GPT-5.5 |
|---|---|---|
| LMArena Text | 1272 | 1471 |
| LMArena Creative Writing | 1269 | 1455 |
| LMArena Multi-Turn | 1267 | 1477 |
| EQ-Bench Creative Writing | — | 1844 |
| EQ-Bench 4 | — | 1315 |
Frequently asked questions
Is GPT-4 Turbo better than GPT-5.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 30.3 on the Noometry Index.
Which is cheaper, GPT-4 Turbo or GPT-5.5?
GPT-5.5 is cheaper. It lists at $5 per million input tokens and $30 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or GPT-5.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.3 versus 33.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 128K.
How many benchmarks do GPT-4 Turbo and GPT-5.5 share?
28 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and GPT-5.5 has 71.