Model comparison
GPT-4 Turbo vs GPT-5 Nano
GPT-5 Nano is the stronger model overall, scoring 33.5 to 30.5 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. GPT-4 Turbo scores higher in 3 categories and GPT-5 Nano in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Nano leads 29.4 to 9.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 81.1% for GPT-5 Nano.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- GPT-5 Nano accepts more context: 400K tokens versus 128K.
Side by side
| GPT-4 Turbo | GPT-5 Nano | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 30.5 | 33.5 |
| Released | 2023-11-06 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 400K |
| Max output | 4K | 128K |
| Input $ / M tokens | $10 | $0.05 |
| Output $ / M tokens | $30 | $0.40 |
| Results tracked | 36 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4 Turbo: 33.8 (#249), GPT-5 Nano: 33.6 (#254)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| WeirdML | 18% | 38.1% |
| LMArena Coding | 1268 | 1351 |
| SWE-bench Verified (bash only) | — | 34.8% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| ALE-Bench | — | 718.67 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, GPT-5 Nano: 25.8 (#106)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |
| METR Time Horizons | 36.7% | — |
Reasoning Too close to call
GPT-4 Turbo: 15.3 (#317), GPT-5 Nano: 16.3 (#306)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| Chess Puzzles | 6% | 27% |
| LMArena Hard Prompts | 1251 | 1328 |
| DTBench | 61.6% | 62.7% |
| LMCA | 9.8% | 7.9% |
| Epoch Capabilities Index | 127.25 | 139.38 |
| ForecastBench | 59.4 | 59.1 |
| ARC-AGI-2 | — | 2.6% |
| SimpleBench | 25.1% | — |
| Kagi LLM Benchmark | — | 62.2% |
| ARC-AGI-1 | — | 20.7% |
| Mystery Game Puzzles | — | 9% |
Math GPT-5 Nano leads
GPT-4 Turbo: 9.0 (#322), GPT-5 Nano: 29.4 (#241)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 20% |
| OTIS Mock AIME 2024-2025 | 6.7% | 81.1% |
| LMArena Math | 1272 | 1317 |
| MATH Level 5 | 46.7% | 95.2% |
| FrontierMath Tier 4 | — | 2.4% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5 Nano leads
GPT-4 Turbo: 24.3 (#268), GPT-5 Nano: 35.9 (#178)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 46.6% | 69.4% |
| LMArena Expert | 1223 | 1321 |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | — | 77.8% |
| Confabulations | 28.4% | — |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |
| MMLU | 81.3% | — |
Multimodal Too close to call
GPT-4 Turbo: 30.6 (#110), GPT-5 Nano: 31.3 (#108)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| LMArena Vision | 1090 | 1159 |
| VPCT | — | 37.2% |
Multilingual GPT-5 Nano leads
GPT-4 Turbo: 40.5 (#216), GPT-5 Nano: 45.3 (#172)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1245 | 1313 |
| LMArena Chinese | 1242 | 1356 |
| LMArena German | 1259 | 1327 |
| LMArena Japanese | 1194 | 1226 |
| LMArena Korean | 1187 | 1269 |
| LMArena Russian | 1259 | 1296 |
| LMArena Spanish | 1260 | 1360 |
| LMArena French | 1276 | — |
Instruction Following GPT-5 Nano leads
GPT-4 Turbo: 65.8 (#216), GPT-5 Nano: 75.0 (#79)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| LMArena Instruction Following | 1249 | 1306 |
| IFEval | — | 93.2% |
Long Context GPT-4 Turbo leads
GPT-4 Turbo: 38.0 (#206), GPT-5 Nano: 31.3 (#281)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| LMArena Longer Query | 1254 | 1312 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference GPT-4 Turbo leads
GPT-4 Turbo: 47.7 (#206), GPT-5 Nano: 39.1 (#249)
| Benchmark | GPT-4 Turbo | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1272 | 1320 |
| LMArena Creative Writing | 1269 | 1249 |
| LMArena Multi-Turn | 1267 | 1311 |
| EQ-Bench Creative Writing | — | 705 |
| WildBench | — | 80.6% |
Frequently asked questions
Is GPT-4 Turbo better than GPT-5 Nano?
GPT-5 Nano is the stronger model overall, scoring 33.5 to 30.5 on the Noometry Index.
Which is cheaper, GPT-4 Turbo or GPT-5 Nano?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; GPT-4 Turbo lists at $10 and $30.
Is GPT-4 Turbo or GPT-5 Nano better for coding?
They score almost the same on coding (33.8 vs 33.6); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 128K.
How many benchmarks do GPT-4 Turbo and GPT-5 Nano share?
27 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and GPT-5 Nano has 49.