Model comparison
Amazon Nova Pro vs GPT-5
GPT-5 is the stronger model overall, scoring 50.9 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.5× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Summary
- They share 28 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and GPT-5 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 38.1.
- The biggest single-benchmark swing is Omni-MATH: 24.2% for Amazon Nova Pro and 64.7% for GPT-5.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $1.25 / $10 for GPT-5.
- GPT-5 accepts more context: 400K tokens versus 300K.
Side by side
| Amazon Nova Pro | GPT-5 | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.0 | 50.9 |
| Released | 2024-12-03 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 400K |
| Max output | 10K | 128K |
| Input $ / M tokens | $0.80 | $1.25 |
| Output $ / M tokens | $3.20 | $10 |
| Results tracked | 38 | 69 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 leads
Amazon Nova Pro: 35.1 (#229), GPT-5: 50.3 (#47)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Coding | 1270 | 1436 |
| SWE-bench Verified | — | 73.6% |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| LMArena WebDev | — | 1418 |
| SciCode | — | 42.9% |
| GSO | — | 6.9% |
| WeirdML | — | 60.7% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 1,162 |
| AlgoTune | — | 1.67 |
Agentic & Tool Use GPT-5 leads
Amazon Nova Pro: 16.7 (#147), GPT-5: 33.1 (#56)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| Terminal-Bench | — | 49.6% |
| Berkeley Function Calling Leaderboard | 25% | — |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| TheAgentCompany | 1.7% | — |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |
Reasoning GPT-5 leads
Amazon Nova Pro: 20.0 (#243), GPT-5: 38.3 (#64)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1416 |
| Epoch Capabilities Index | 123.8 | 150 |
| ARC-AGI-2 | — | 9.9% |
| SimpleBench | — | 56.7% |
| Kagi LLM Benchmark | — | 72.7% |
| ARC-AGI-1 | — | 65.7% |
| CritPt | — | 12.6% |
| Chess Puzzles | — | 37% |
| EnigmaEval | — | 10.5% |
| EBR-Bench | — | 12.7% |
| LiveBench Reasoning | 32.6% | — |
| Mystery Game Puzzles | — | 23% |
| DTBench | — | 90.7% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 40% |
| ForecastBench | — | 61.4 |
| LiveBench | 43.5% | — |
Math GPT-5 leads
Amazon Nova Pro: 28.5 (#243), GPT-5: 55.0 (#44)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| Omni-MATH | 24.2% | 64.7% |
| LMArena Math | 1252 | 1407 |
| FrontierMath (Tiers 1-3) | — | 55.4% |
| FrontierMath Tier 4 | — | 22% |
| OTIS Mock AIME 2024-2025 | — | 91.4% |
| ProofBench | — | 18% |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GPT-5 leads
Amazon Nova Pro: 27.4 (#250), GPT-5: 56.6 (#43)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| Humanity's Last Exam | 4.4% | 25.3% |
| MMLU-Pro | 67.3% | 86.3% |
| Confabulations | 30.1% | 10.3% |
| Vectara Hallucination Rate | 5.1% | 14.7% |
| GPQA (HELM) | 44.6% | 79.2% |
| LMArena Expert | 1211 | 1419 |
| GPQA Diamond | — | 86.2% |
| SimpleQA Verified | — | 50.1% |
| MMLU | 82% | — |
Multimodal GPT-5 leads
Amazon Nova Pro: 25.0 (#126), GPT-5: 46.8 (#13)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Vision | 980 | 1232 |
| GeoBench | — | 81% |
| VPCT | — | 66% |
Multilingual GPT-5 leads
Amazon Nova Pro: 39.7 (#223), GPT-5: 51.4 (#110)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Non-English | 1234 | 1397 |
| LMArena Chinese | 1244 | 1422 |
| LMArena French | 1271 | 1410 |
| LMArena German | 1243 | 1416 |
| LMArena Japanese | 1200 | 1409 |
| LMArena Korean | 1203 | 1360 |
| LMArena Russian | 1240 | 1406 |
| LMArena Spanish | 1182 | 1399 |
Instruction Following GPT-5 leads
Amazon Nova Pro: 64.9 (#226), GPT-5: 73.8 (#113)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| IFEval | 81.5% | 87.5% |
| LMArena Instruction Following | 1235 | 1388 |
| LiveBench Instruction Following | 67.1% | — |
Long Context GPT-5 leads
Amazon Nova Pro: 38.1 (#205), GPT-5: 69.5 (#2)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1255 | 1399 |
| Fiction.LiveBench | — | 97.2% |
Writing & Preference GPT-5 leads
Amazon Nova Pro: 43.9 (#226), GPT-5: 63.4 (#65)
| Benchmark | Amazon Nova Pro | GPT-5 |
|---|---|---|
| LMArena Text | 1259 | 1406 |
| LMArena Creative Writing | 1212 | 1365 |
| Short-Story Creative Writing | 60.5% | 86% |
| WildBench | 77.7% | 85.7% |
| LMArena Multi-Turn | 1246 | 1426 |
| EQ-Bench Creative Writing | — | 1627 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than GPT-5?
GPT-5 is the stronger model overall, scoring 50.9 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.5× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Pro or GPT-5?
Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; GPT-5 lists at $1.25 and $10.
Is Amazon Nova Pro or GPT-5 better for coding?
GPT-5 scores higher on coding benchmarks: 50.3 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5 does, with 400K tokens against 300K.
How many benchmarks do Amazon Nova Pro and GPT-5 share?
28 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and GPT-5 has 69.