Model comparison
GPT-5.1 vs GPT-5 Mini
GPT-5.1 is the stronger model overall, scoring 48.8 to 41.7 on the Noometry Index. GPT-5 Mini costs 5.0× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 48 shared benchmarks.
Summary
- They share 48 benchmarks with published results for both. GPT-5.1 scores higher in 10 categories and GPT-5 Mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.1 leads 39.9 to 24.1.
- The biggest single-benchmark swing is GPQA (HELM): 44.2% for GPT-5.1 and 75.6% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
Side by side
| GPT-5.1 | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 48.8 | 41.7 |
| Released | 2025-11-13 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $1.25 | $0.25 |
| Output $ / M tokens | $10 | $2 |
| Results tracked | 63 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-5.1: 46.9 (#59), GPT-5 Mini: 40.0 (#134)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 68% | 64.7% |
| SWE-bench Verified (bash only) | 66% | 59.8% |
| SciCode | 43.3% | 39.2% |
| WeirdML | 60.8% | 52.7% |
| LMArena Coding | 1454 | 1406 |
| ALE-Bench | 1,192 | 799.77 |
| LMArena WebDev | 1395 | — |
| SWE-bench Multilingual | — | 39.7% |
| GSO | 13.7% | — |
| LiveBench Coding | 72.5% | — |
| AlgoTune | — | 1.38 |
Agentic & Tool Use GPT-5.1 leads
GPT-5.1: 32.5 (#55), GPT-5 Mini: 31.0 (#64)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | 47.6% | 34.8% |
| Vending-Bench 2 | 1,473 | -31.18 |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
Reasoning GPT-5.1 leads
GPT-5.1: 39.9 (#50), GPT-5 Mini: 24.1 (#153)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 17.6% | 4.4% |
| ARC-AGI-1 | 72.8% | 54.3% |
| CritPt | 4.9% | 0% |
| Chess Puzzles | 32% | 30% |
| EnigmaEval | 11.2% | 8.2% |
| LMArena Hard Prompts | 1457 | 1380 |
| Mystery Game Puzzles | 19% | 10% |
| DTBench | 90.1% | 80.5% |
| LMCA | 43.9% | 34.2% |
| Epoch Capabilities Index | 149.66 | 145.51 |
| ForecastBench | 58.1 | 61 |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | — | 70.3% |
| LiveBench Reasoning | 95.8% | — |
| LiveBench Data Analysis | 72.1% | — |
| LiveBench | 78.8% | — |
Math GPT-5.1 leads
GPT-5.1: 50.3 (#45), GPT-5 Mini: 46.3 (#59)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 86.7% |
| Omni-MATH | 46.4% | 72.2% |
| LMArena Math | 1446 | 1377 |
| FrontierMath (Feb 2025 set) | 31% | 27.2% |
| FrontierMath Tier 4 (v1) | 12.5% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| LiveBench Math | 94.5% | — |
| MATH Level 5 | — | 97.8% |
Knowledge GPT-5.1 leads
GPT-5.1: 50.5 (#65), GPT-5 Mini: 45.5 (#80)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 87.6% | 75% |
| Humanity's Last Exam | 23.7% | 19.4% |
| SimpleQA Verified | 48% | 21.6% |
| MMLU-Pro | 57.9% | 83.5% |
| Vectara Hallucination Rate | 10.9% | 12.9% |
| GPQA (HELM) | 44.2% | 75.6% |
| LMArena Expert | 1470 | 1377 |
| Confabulations | — | 13.3% |
Multimodal GPT-5.1 leads
GPT-5.1: 44.0 (#17), GPT-5 Mini: 35.0 (#80)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1250 | 1202 |
| VPCT | 58.7% | 40.2% |
| LMArena Document | 1403 | — |
Multilingual GPT-5.1 leads
GPT-5.1: 54.0 (#51), GPT-5 Mini: 49.1 (#131)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1431 | 1363 |
| LMArena Chinese | 1495 | 1387 |
| LMArena French | 1451 | 1385 |
| LMArena German | 1439 | 1366 |
| LMArena Japanese | 1453 | 1342 |
| LMArena Korean | 1401 | 1306 |
| LMArena Russian | 1435 | 1362 |
| LMArena Spanish | 1430 | 1355 |
Instruction Following GPT-5.1 leads
GPT-5.1: 84.0 (#1), GPT-5 Mini: 76.4 (#43)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| IFEval | 93.5% | 92.7% |
| LMArena Instruction Following | 1442 | 1357 |
| LiveBench Instruction Following | 93.3% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), GPT-5 Mini: 41.9 (#127)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1448 | 1355 |
| Fiction.LiveBench | — | 69.4% |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.7 (#53), GPT-5 Mini: 55.4 (#143)
| Benchmark | GPT-5.1 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1443 | 1373 |
| LMArena Creative Writing | 1427 | 1325 |
| WildBench | 86.3% | 85.5% |
| LMArena Multi-Turn | 1449 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| EQ-Bench Creative Writing | — | 1313 |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than GPT-5 Mini?
GPT-5.1 is the stronger model overall, scoring 48.8 to 41.7 on the Noometry Index. GPT-5 Mini costs 5.0× 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-5.1 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-5.1 lists at $1.25 and $10.
Is GPT-5.1 or GPT-5 Mini better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.9 versus 40.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 400K tokens.
How many benchmarks do GPT-5.1 and GPT-5 Mini share?
48 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and GPT-5 Mini has 60.