Model comparison
GPT-4o mini vs GPT-5.1
GPT-5.1 is the stronger model overall, scoring 49.0 to 25.5 on the Noometry Index. GPT-4o mini costs 13× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 42 shared benchmarks.
Summary
- They share 42 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and GPT-5.1 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.1 leads 52.2 to 10.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 88.6% for GPT-5.1.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- GPT-5.1 accepts more context: 400K tokens versus 128K.
Side by side
| GPT-4o mini | GPT-5.1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 25.5 | 49.0 |
| Released | 2024-07-18 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 400K |
| Max output | 16K | 128K |
| Input $ / M tokens | $0.15 | $1.25 |
| Output $ / M tokens | $0.60 | $10 |
| Results tracked | 60 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-4o mini: 22.0 (#335), GPT-5.1: 46.4 (#66)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| WeirdML | 11.8% | 60.8% |
| LiveBench Coding | 43.1% | 72.5% |
| LMArena Coding | 1290 | 1454 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| Aider Polyglot | 3.6% | — |
| LMArena WebDev | — | 1395 |
| SciCode | — | 43.3% |
| GSO | — | 13.7% |
| BigCodeBench Instruct | 46.1% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 1,192 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use GPT-5.1 leads
GPT-4o mini: 27.5 (#101), GPT-5.1: 32.7 (#60)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| DeepResearch Bench | — | 42.8% |
| BALROG | 17.4% | — |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |
Reasoning GPT-5.1 leads
GPT-4o mini: 8.7 (#347), GPT-5.1: 39.8 (#58)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 0% | 17.6% |
| SimpleBench | 10.7% | 53.2% |
| Chess Puzzles | 0% | 32% |
| LiveBench Reasoning | 32.8% | 95.8% |
| LMArena Hard Prompts | 1267 | 1457 |
| Mystery Game Puzzles | 12% | 19% |
| DTBench | 54.4% | 90.1% |
| LiveBench Data Analysis | 50% | 72.1% |
| LMCA | 10.4% | 43.9% |
| Epoch Capabilities Index | 126.56 | 149.64 |
| LiveBench | 41.3% | 78.8% |
| Kagi LLM Benchmark | 28.8% | — |
| ARC-AGI-1 | — | 72.8% |
| CritPt | — | 4.9% |
| EnigmaEval | — | 11.2% |
| ForecastBench | — | 58.1 |
| PIQA | 88.7% | — |
Math GPT-5.1 leads
GPT-4o mini: 10.4 (#314), GPT-5.1: 52.2 (#51)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.9% | 88.6% |
| Omni-MATH | 28% | 46.4% |
| LiveBench Math | 36.3% | 94.5% |
| LMArena Math | 1267 | 1447 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| MATH Level 5 | 52.6% | — |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
| GSM8K | 91.3% | — |
Knowledge GPT-5.1 leads
GPT-4o mini: 17.7 (#284), GPT-5.1: 50.6 (#71)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 37.7% | 87.6% |
| SimpleQA Verified | 8.3% | 48% |
| MMLU-Pro | 60.3% | 57.9% |
| GPQA (HELM) | 36.8% | 44.2% |
| LMArena Expert | 1235 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| Confabulations | 37.2% | — |
| Vectara Hallucination Rate | — | 10.9% |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |
Multimodal GPT-5.1 leads
GPT-4o mini: 25.9 (#122), GPT-5.1: 44.8 (#19)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1066 | 1250 |
| VPCT | 34% | 58.7% |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| LMArena Document | — | 1403 |
Multilingual GPT-5.1 leads
GPT-4o mini: 42.0 (#199), GPT-5.1: 53.8 (#56)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1266 | 1431 |
| LMArena Chinese | 1265 | 1495 |
| LMArena French | 1297 | 1450 |
| LMArena German | 1272 | 1438 |
| LMArena Japanese | 1216 | 1453 |
| LMArena Korean | 1195 | 1401 |
| LMArena Russian | 1275 | 1435 |
| LMArena Spanish | 1276 | 1433 |
Instruction Following GPT-5.1 leads
GPT-4o mini: 61.9 (#239), GPT-5.1: 83.9 (#1)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 56.8% | 93.3% |
| IFEval | 78.2% | 93.5% |
| LMArena Instruction Following | 1258 | 1443 |
Long Context GPT-5.1 leads
GPT-4o mini: 39.1 (#186), GPT-5.1: 47.6 (#14)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1289 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference GPT-5.1 leads
GPT-4o mini: 39.5 (#248), GPT-5.1: 64.5 (#55)
| Benchmark | GPT-4o mini | GPT-5.1 |
|---|---|---|
| LMArena Text | 1286 | 1443 |
| LMArena Creative Writing | 1268 | 1427 |
| WildBench | 79.1% | 86.3% |
| LMArena Multi-Turn | 1285 | 1450 |
| LiveBench Language | 28.6% | 80.2% |
| Short-Story Creative Writing | 67.2% | — |
| EQ-Bench Creative Writing | 873 | — |
Frequently asked questions
Is GPT-4o mini better than GPT-5.1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 25.5 on the Noometry Index. GPT-4o mini costs 13× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Which is cheaper, GPT-4o mini or GPT-5.1?
GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is GPT-4o mini or GPT-5.1 better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 22.0 in the Noometry coding category.
Which has the bigger context window?
GPT-5.1 does, with 400K tokens against 128K.
How many benchmarks do GPT-4o mini and GPT-5.1 share?
42 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and GPT-5.1 has 63.