Model comparison
GPT-3.5-turbo vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.8 to 23.2 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and GPT-5 Mini in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.7 to 6.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.2% for GPT-3.5-turbo and 86.7% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
- GPT-5 Mini accepts more context: 400K tokens versus 16K.
Side by side
| GPT-3.5-turbo | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 23.2 | 41.8 |
| Released | 2023-03-01 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 16K | 400K |
| Max output | 4K | 128K |
| Input $ / M tokens | $0.50 | $0.25 |
| Output $ / M tokens | $1.50 | $2 |
| Results tracked | 44 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-3.5-turbo: 23.9 (#331), GPT-5 Mini: 40.1 (#146)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| WeirdML | 3.5% | 52.7% |
| LMArena Coding | 1136 | 1406 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| BigCodeBench Instruct | 39.1% | — |
| BigCodeBench Complete | 50.6% | — |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, GPT-5 Mini: 31.1 (#70)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| METR Time Horizons | 21.5% | — |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
GPT-3.5-turbo: 13.8 (#332), GPT-5 Mini: 23.9 (#168)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| Chess Puzzles | 0% | 30% |
| LMArena Hard Prompts | 1108 | 1380 |
| Mystery Game Puzzles | 3% | 10% |
| DTBench | 48.5% | 80.5% |
| LMCA | 9.7% | 34.2% |
| Epoch Capabilities Index | 118.55 | 145.52 |
| ForecastBench | 50.4 | 61 |
| ARC-AGI-2 | — | 4.4% |
| Kagi LLM Benchmark | — | 70.3% |
| ARC-AGI-1 | — | 54.3% |
| CritPt | — | 0% |
| EnigmaEval | — | 8.2% |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| WinoGrande | 81.6% | — |
Math GPT-5 Mini leads
GPT-3.5-turbo: 6.3 (#327), GPT-5 Mini: 46.7 (#69)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0% | 46.7% |
| OTIS Mock AIME 2024-2025 | 2.2% | 86.7% |
| LMArena Math | 1142 | 1378 |
| MATH Level 5 | 15.9% | 97.8% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
| GSM8K | 57.8% | — |
Knowledge GPT-5 Mini leads
GPT-3.5-turbo: 10.0 (#303), GPT-5 Mini: 45.6 (#86)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 28% | 75% |
| LMArena Expert | 1070 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multimodal Not comparable
GPT-3.5-turbo: —, GPT-5 Mini: 35.6 (#85)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual GPT-5 Mini leads
GPT-3.5-turbo: 31.5 (#258), GPT-5 Mini: 48.9 (#137)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1108 | 1363 |
| LMArena Chinese | 1075 | 1385 |
| LMArena French | 1118 | 1386 |
| LMArena German | 1090 | 1366 |
| LMArena Japanese | 1043 | 1341 |
| LMArena Korean | 1019 | 1308 |
| LMArena Russian | 1123 | 1362 |
| LMArena Spanish | 1121 | 1355 |
Instruction Following GPT-5 Mini leads
GPT-3.5-turbo: 57.9 (#262), GPT-5 Mini: 76.2 (#46)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1119 | 1357 |
| IFEval | — | 92.7% |
Long Context GPT-5 Mini leads
GPT-3.5-turbo: 34.0 (#254), GPT-5 Mini: 41.9 (#132)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1121 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference GPT-5 Mini leads
GPT-3.5-turbo: 25.3 (#305), GPT-5 Mini: 55.2 (#148)
| Benchmark | GPT-3.5-turbo | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1125 | 1373 |
| LMArena Creative Writing | 1092 | 1325 |
| EQ-Bench Creative Writing | 451 | 1313 |
| LMArena Multi-Turn | 1117 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
Frequently asked questions
Is GPT-3.5-turbo better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.8 to 23.2 on the Noometry Index.
Which is cheaper, GPT-3.5-turbo or GPT-5 Mini?
GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.
Is GPT-3.5-turbo or GPT-5 Mini better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.1 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and GPT-5 Mini share?
29 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and GPT-5 Mini has 60.