Model comparison
GPT-3.5-turbo vs o3-pro
o3-pro is the stronger model overall, scoring 42.7 to 23.1 on the Noometry Index. GPT-3.5-turbo costs 47× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Summary
- They share 4 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and o3-pro in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.1 to 34.0.
- The biggest single-benchmark swing is WeirdML: 3.5% for GPT-3.5-turbo and 58.2% for o3-pro.
- GPT-3.5-turbo is cheaper at $0.50 / $1.50 per million input/output tokens, against $20 / $80 for o3-pro.
- o3-pro accepts more context: 200K tokens versus 16K.
Side by side
| GPT-3.5-turbo | o3-pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 23.1 | 42.7 |
| Released | 2023-03-01 | 2025-06-10 |
| Weights | Proprietary | Proprietary |
| Context window | 16K | 200K |
| Max output | 4K | 100K |
| Input $ / M tokens | $0.50 | $20 |
| Output $ / M tokens | $1.50 | $80 |
| Results tracked | 44 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
GPT-3.5-turbo: 23.5 (#326), o3-pro: 55.1 (#22)
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| WeirdML | 3.5% | 58.2% |
| Aider Polyglot | — | 84.9% |
| BigCodeBench Instruct | 39.1% | — |
| LMArena Coding | 1137 | — |
| BigCodeBench Complete | 50.6% | — |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, o3-pro: —
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| METR Time Horizons | 21.5% | — |
Reasoning o3-pro leads
GPT-3.5-turbo: 13.3 (#328), o3-pro: 24.1 (#152)
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| DTBench | 48.5% | 86.9% |
| LMCA | 9.7% | 38.5% |
| Epoch Capabilities Index | 118.48 | 147.43 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 72.1% |
| ARC-AGI-1 | — | 59.3% |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | 1108 | — |
| Mystery Game Puzzles | 3% | — |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| ForecastBench | 50.4 | — |
| WinoGrande | 81.6% | — |
Math Not comparable
GPT-3.5-turbo: 6.0 (#321), o3-pro: —
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0% | — |
| OTIS Mock AIME 2024-2025 | 2.2% | — |
| LMArena Math | 1142 | — |
| MATH Level 5 | 15.9% | — |
| GSM8K | 57.8% | — |
Knowledge o3-pro leads
GPT-3.5-turbo: 10.0 (#298), o3-pro: 29.0 (#234)
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| GPQA Diamond | 28% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1070 | — |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual Not comparable
GPT-3.5-turbo: 31.6 (#253), o3-pro: —
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| LMArena Non-English | 1108 | — |
| LMArena Chinese | 1075 | — |
| LMArena French | 1117 | — |
| LMArena German | 1090 | — |
| LMArena Japanese | 1043 | — |
| LMArena Korean | 1019 | — |
| LMArena Russian | 1123 | — |
| LMArena Spanish | 1120 | — |
Instruction Following Not comparable
GPT-3.5-turbo: 58.1 (#257), o3-pro: —
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1119 | — |
Long Context o3-pro leads
GPT-3.5-turbo: 34.0 (#249), o3-pro: 72.1 (#1)
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1121 | — |
Writing & Preference o3-pro leads
GPT-3.5-turbo: 25.4 (#300), o3-pro: 57.1 (#128)
| Benchmark | GPT-3.5-turbo | o3-pro |
|---|---|---|
| LMArena Text | 1125 | — |
| LMArena Creative Writing | 1092 | — |
| Short-Story Creative Writing | — | 84.4% |
| EQ-Bench Creative Writing | 451 | — |
| LMArena Multi-Turn | 1117 | — |
Frequently asked questions
Is GPT-3.5-turbo better than o3-pro?
o3-pro is the stronger model overall, scoring 42.7 to 23.1 on the Noometry Index. GPT-3.5-turbo costs 47× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.
Which is cheaper, GPT-3.5-turbo or o3-pro?
GPT-3.5-turbo is cheaper. It lists at $0.50 per million input tokens and $1.50 per million output tokens; o3-pro lists at $20 and $80.
Is GPT-3.5-turbo or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.1 versus 23.5 in the Noometry coding category.
Which has the bigger context window?
o3-pro does, with 200K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and o3-pro share?
4 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and o3-pro has 12.