Model comparison
GPT-4 Turbo vs GPT-4o
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 28.5 on the Noometry Index. GPT-4o costs 3.4× less per token, which makes it the better buy when GPT-4 Turbo's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-4 Turbo scores higher in 2 categories and GPT-4o in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-4 Turbo leads 33.4 to 25.1.
- The biggest single-benchmark swing is Confabulations: 28.4% for GPT-4 Turbo and 15.3% for GPT-4o.
- GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
Side by side
| GPT-4 Turbo | GPT-4o | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 30.3 | 28.5 |
| Released | 2023-11-06 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 128K |
| Max output | 4K | 16K |
| Input $ / M tokens | $10 | $2.50 |
| Output $ / M tokens | $30 | $10 |
| Results tracked | 36 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
GPT-4 Turbo: 33.4 (#247), GPT-4o: 25.1 (#323)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| WeirdML | 18% | 25.1% |
| BigCodeBench Instruct | 48.2% | 51.1% |
| LMArena Coding | 1268 | 1297 |
| BigCodeBench Complete | 58.2% | 61.1% |
| HumanEval+ | 86.6% | 87.2% |
| MBPP+ | 73.3% | 72.2% |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
| Aider Polyglot | — | 45.3% |
| GSO | — | 0% |
| LiveBench Coding | — | 51.4% |
| CadEval | — | 26% |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, GPT-4o: 20.8 (#133)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| METR Time Horizons | 36.7% | 40.8% |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
Reasoning GPT-4 Turbo leads
GPT-4 Turbo: 14.9 (#311), GPT-4o: 9.4 (#338)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| SimpleBench | 25.1% | 17.8% |
| Chess Puzzles | 6% | 13% |
| LMArena Hard Prompts | 1251 | 1281 |
| DTBench | 61.6% | 64.5% |
| LMCA | 9.8% | 16.6% |
| Epoch Capabilities Index | 127.25 | 128.97 |
| ForecastBench | 59.4 | 57.7 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 4.5% |
| CritPt | — | 0% |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| LiveBench Data Analysis | — | 60.9% |
| LiveBench | — | 55.3% |
Math GPT-4o leads
GPT-4 Turbo: 8.6 (#316), GPT-4o: 10.2 (#303)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 0.4% |
| OTIS Mock AIME 2024-2025 | 6.7% | 6.4% |
| LMArena Math | 1272 | 1285 |
| MATH Level 5 | 46.7% | 53.3% |
| Omni-MATH | — | 29.3% |
| LiveBench Math | — | 49.5% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GPT-4o leads
GPT-4 Turbo: 24.3 (#263), GPT-4o: 28.8 (#235)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| GPQA Diamond | 46.6% | 49.2% |
| Confabulations | 28.4% | 15.3% |
| LMArena Expert | 1223 | 1250 |
| MMLU | 81.3% | 88.1% |
| Humanity's Last Exam | — | 2.7% |
| SimpleQA Verified | — | 26% |
| MMLU-Pro | — | 71.3% |
| Vectara Hallucination Rate | — | 9.6% |
| GPQA (HELM) | — | 52% |
Multimodal GPT-4o leads
GPT-4 Turbo: 30.0 (#104), GPT-4o: 33.9 (#85)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| LMArena Vision | 1090 | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |
Multilingual GPT-4o leads
GPT-4 Turbo: 40.7 (#211), GPT-4o: 43.4 (#181)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| LMArena Non-English | 1245 | 1283 |
| LMArena Chinese | 1242 | 1277 |
| LMArena French | 1275 | 1303 |
| LMArena German | 1259 | 1282 |
| LMArena Japanese | 1194 | 1257 |
| LMArena Korean | 1187 | 1234 |
| LMArena Russian | 1259 | 1286 |
| LMArena Spanish | 1260 | 1292 |
Instruction Following Too close to call
GPT-4 Turbo: 66.0 (#211), GPT-4o: 66.8 (#202)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| LMArena Instruction Following | 1250 | 1278 |
| LiveBench Instruction Following | — | 68.6% |
| IFEval | — | 81.7% |
Long Context GPT-4o leads
GPT-4 Turbo: 38.0 (#200), GPT-4o: 39.4 (#174)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1254 | 1289 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference GPT-4o leads
GPT-4 Turbo: 47.9 (#201), GPT-4o: 52.8 (#160)
| Benchmark | GPT-4 Turbo | GPT-4o |
|---|---|---|
| LMArena Text | 1272 | 1300 |
| LMArena Creative Writing | 1269 | 1291 |
| LMArena Multi-Turn | 1267 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| WildBench | — | 82.8% |
| LiveBench Language | — | 47.6% |
Frequently asked questions
Is GPT-4 Turbo better than GPT-4o?
GPT-4 Turbo is the stronger model overall, scoring 30.3 to 28.5 on the Noometry Index. GPT-4o costs 3.4× less per token, which makes it the better buy when GPT-4 Turbo's lead doesn't matter for your workload.
Which is cheaper, GPT-4 Turbo or GPT-4o?
GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or GPT-4o better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.4 versus 25.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do GPT-4 Turbo and GPT-4o share?
36 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and GPT-4o has 72.