Model comparison
GPT-5 Mini vs GPT-5 Nano
GPT-5 Mini is the stronger model overall, scoring 41.7 to 33.4 on the Noometry Index. GPT-5 Nano costs 5.0× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Last verified . 48 shared benchmarks.
Summary
- They share 48 benchmarks with published results for both. GPT-5 Mini scores higher in 10 categories and GPT-5 Nano in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 29.1.
- The biggest single-benchmark swing is ARC-AGI-1: 54.3% for GPT-5 Mini and 20.7% for GPT-5 Nano.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
Side by side
| GPT-5 Mini | GPT-5 Nano | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 41.7 | 33.4 |
| Released | 2025-08-07 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.25 | $0.05 |
| Output $ / M tokens | $2 | $0.40 |
| Results tracked | 60 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.0 (#134), GPT-5 Nano: 33.3 (#248)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| SWE-bench Verified (bash only) | 59.8% | 34.8% |
| WeirdML | 52.7% | 38.1% |
| LMArena Coding | 1406 | 1351 |
| ALE-Bench | 799.77 | 718.67 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.0 (#64), GPT-5 Nano: 25.6 (#99)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | 34.8% | 21.8% |
| Berkeley Function Calling Leaderboard | 55.5% | 51.5% |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 24.1 (#153), GPT-5 Nano: 16.4 (#291)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| ARC-AGI-2 | 4.4% | 2.6% |
| Kagi LLM Benchmark | 70.3% | 62.2% |
| ARC-AGI-1 | 54.3% | 20.7% |
| Chess Puzzles | 30% | 27% |
| LMArena Hard Prompts | 1380 | 1329 |
| Mystery Game Puzzles | 10% | 9% |
| DTBench | 80.5% | 62.7% |
| LMCA | 34.2% | 7.9% |
| Epoch Capabilities Index | 145.51 | 139.39 |
| ForecastBench | 61 | 59.1 |
| CritPt | 0% | — |
| EnigmaEval | 8.2% | — |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), GPT-5 Nano: 29.1 (#221)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| FrontierMath (Tiers 1-3) | 46.7% | 20% |
| FrontierMath Tier 4 | 12.2% | 2.4% |
| OTIS Mock AIME 2024-2025 | 86.7% | 81.1% |
| ProofBench | 9% | 12% |
| Omni-MATH | 72.2% | 54.6% |
| LMArena Math | 1377 | 1318 |
| MATH Level 5 | 97.8% | 95.2% |
| FrontierMath (Feb 2025 set) | 27.2% | 8.3% |
| FrontierMath Tier 4 (v1) | 6.3% | 2.1% |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.5 (#80), GPT-5 Nano: 35.8 (#166)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 75% | 69.4% |
| SimpleQA Verified | 21.6% | 11.7% |
| MMLU-Pro | 83.5% | 77.8% |
| Vectara Hallucination Rate | 12.9% | 10.5% |
| GPQA (HELM) | 75.6% | 67.9% |
| LMArena Expert | 1377 | 1321 |
| Humanity's Last Exam | 19.4% | — |
| Confabulations | 13.3% | — |
Multimodal GPT-5 Mini leads
GPT-5 Mini: 35.0 (#80), GPT-5 Nano: 30.7 (#102)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| LMArena Vision | 1202 | 1159 |
| VPCT | 40.2% | 37.2% |
Multilingual GPT-5 Mini leads
GPT-5 Mini: 49.1 (#131), GPT-5 Nano: 45.5 (#167)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1363 | 1313 |
| LMArena Chinese | 1387 | 1355 |
| LMArena German | 1366 | 1329 |
| LMArena Korean | 1306 | 1269 |
| LMArena Russian | 1362 | 1296 |
| LMArena Spanish | 1355 | 1360 |
| LMArena French | 1385 | — |
| LMArena Japanese | 1342 | — |
Instruction Following GPT-5 Mini leads
GPT-5 Mini: 76.4 (#43), GPT-5 Nano: 75.2 (#72)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| IFEval | 92.7% | 93.2% |
| LMArena Instruction Following | 1357 | 1306 |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#127), GPT-5 Nano: 31.3 (#275)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| Fiction.LiveBench | 69.4% | 44.4% |
| LMArena Longer Query | 1355 | 1313 |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.4 (#143), GPT-5 Nano: 39.4 (#244)
| Benchmark | GPT-5 Mini | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1373 | 1320 |
| LMArena Creative Writing | 1325 | 1250 |
| EQ-Bench Creative Writing | 1313 | 705 |
| WildBench | 85.5% | 80.6% |
| LMArena Multi-Turn | 1363 | 1311 |
| Short-Story Creative Writing | 83.1% | — |
Frequently asked questions
Is GPT-5 Mini better than GPT-5 Nano?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 33.4 on the Noometry Index. GPT-5 Nano costs 5.0× less per token, which makes it the better buy when GPT-5 Mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Mini 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-5 Mini lists at $0.25 and $2.
Is GPT-5 Mini or GPT-5 Nano better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.0 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
Both accept 400K tokens.
How many benchmarks do GPT-5 Mini and GPT-5 Nano share?
48 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and GPT-5 Nano has 48.