Model comparison
Claude Opus 5 vs GPT-3.5-turbo
Claude Opus 5 is the stronger model overall, scoring 67.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 13× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude Opus 5 scores higher in 8 categories and GPT-3.5-turbo in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5 leads 86.2 to 6.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 98.9% for Claude Opus 5 and 2.2% for GPT-3.5-turbo.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 16K.
Side by side
| Claude Opus 5 | GPT-3.5-turbo | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 67.8 | 23.2 |
| Released | 2026-07-24 | 2023-03-01 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 16K |
| Max output | 128K | 4K |
| Input $ / M tokens | $5 | $0.50 |
| Output $ / M tokens | $25 | $1.50 |
| Results tracked | 57 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.5 (#5), GPT-3.5-turbo: 23.9 (#331)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| WeirdML | 91.8% | 3.5% |
| LMArena Coding | 1534 | 1136 |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| CursorBench | 46.6% | — |
| LMArena WebDev | 1691 | — |
| FrontierSWE | 52% | — |
| SciCode | 56.4% | — |
| BigCodeBench Instruct | — | 39.1% |
| BigCodeBench Complete | — | 50.6% |
| ALE-Bench | 2,165 | — |
| HumanEval+ | — | 70.7% |
| MBPP+ | — | 69.7% |
Agentic & Tool Use Not comparable
Claude Opus 5: 55.6 (#1), GPT-3.5-turbo: —
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| METR Time Horizons | — | 21.5% |
| Vending-Bench 2 | 11,182 | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.2 (#4), GPT-3.5-turbo: 13.8 (#332)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| Chess Puzzles | 42% | 0% |
| LMArena Hard Prompts | 1526 | 1108 |
| Mystery Game Puzzles | 59% | 3% |
| DTBench | 97.9% | 48.5% |
| LMCA | 64.5% | 9.7% |
| Epoch Capabilities Index | 162.78 | 118.55 |
| ARC-AGI-2 | 90.4% | — |
| SimpleBench | 80.6% | — |
| NYT Connections (extended) | 94.3% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 29.1% | — |
| EBR-Bench | 45.7% | — |
| Adversarial NLI | — | 58.1% |
| Bench to the Future 3 | 0.12 | — |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| ForecastBench | — | 50.4 |
| WinoGrande | — | 81.6% |
Math Claude Opus 5 leads
Claude Opus 5: 86.2 (#8), GPT-3.5-turbo: 6.3 (#327)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 0% |
| OTIS Mock AIME 2024-2025 | 98.9% | 2.2% |
| LMArena Math | 1531 | 1142 |
| FrontierMath Tier 4 | 73.2% | — |
| ProofBench | 99% | — |
| MATH Level 5 | — | 15.9% |
| GSM8K | — | 57.8% |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#9), GPT-3.5-turbo: 10.0 (#303)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| GPQA Diamond | 93.9% | 28% |
| LMArena Expert | 1557 | 1070 |
| SimpleQA Verified | 59.9% | — |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| MMLU | — | 71.4% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multimodal Not comparable
Claude Opus 5: 50.8 (#8), GPT-3.5-turbo: —
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| LMArena Vision | 1319 | — |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#4), GPT-3.5-turbo: 31.5 (#258)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1501 | 1108 |
| LMArena Chinese | 1574 | 1075 |
| LMArena French | 1519 | 1118 |
| LMArena German | 1524 | 1090 |
| LMArena Japanese | 1516 | 1043 |
| LMArena Korean | 1521 | 1019 |
| LMArena Russian | 1507 | 1123 |
| LMArena Spanish | 1519 | 1121 |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.2 (#7), GPT-3.5-turbo: 57.9 (#262)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1517 | 1119 |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#21), GPT-3.5-turbo: 34.0 (#254)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1515 | 1121 |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.2 (#1), GPT-3.5-turbo: 25.3 (#305)
| Benchmark | Claude Opus 5 | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1507 | 1125 |
| LMArena Creative Writing | 1491 | 1092 |
| EQ-Bench Creative Writing | 2133 | 451 |
| LMArena Multi-Turn | 1499 | 1117 |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than GPT-3.5-turbo?
Claude Opus 5 is the stronger model overall, scoring 67.8 to 23.2 on the Noometry Index. GPT-3.5-turbo costs 13× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or GPT-3.5-turbo?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or GPT-3.5-turbo better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.5 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 5 does, with 1M tokens against 16K.
How many benchmarks do Claude Opus 5 and GPT-3.5-turbo share?
27 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GPT-3.5-turbo has 44.