Model comparison
GPT-5.6 Terra vs GPT-5 Mini
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 41.8 on the Noometry Index. GPT-5 Mini costs 6.5× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. GPT-5.6 Terra scores higher in 10 categories and GPT-5 Mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Terra leads 60.7 to 23.9.
- The biggest single-benchmark swing is ARC-AGI-2: 83.9% for GPT-5.6 Terra and 4.4% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5.6 Terra | GPT-5 Mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 59.2 | 41.8 |
| Released | 2026-07-09 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $0.25 |
| Output $ / M tokens | $12 | $2 |
| Results tracked | 52 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
GPT-5.6 Terra: 57.7 (#19), GPT-5 Mini: 40.1 (#146)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| SciCode | 55% | 39.2% |
| WeirdML | 78.3% | 52.7% |
| LMArena Coding | 1484 | 1406 |
| ALE-Bench | 1,951 | 799.77 |
| SWE-bench Verified | — | 64.7% |
| DeepSWE | 69.6% | — |
| FrontierCode | 41.3% | — |
| SWE-bench Verified (bash only) | — | 59.8% |
| CursorBench | 41.3% | — |
| LMArena WebDev | 1522 | — |
| SWE-bench Multilingual | — | 39.7% |
| AlgoTune | — | 1.38 |
Agentic & Tool Use GPT-5.6 Terra leads
GPT-5.6 Terra: 40.1 (#25), GPT-5 Mini: 31.1 (#70)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| Vending-Bench 2 | 7,343 | -31.18 |
| Terminal-Bench | — | 34.8% |
| APEX-Agents | 58.2% | — |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| BALROG | 53.2% | — |
| GDP.pdf | 24.7% | — |
Reasoning GPT-5.6 Terra leads
GPT-5.6 Terra: 60.7 (#21), GPT-5 Mini: 23.9 (#168)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 83.9% | 4.4% |
| Kagi LLM Benchmark | 51.3% | 70.3% |
| ARC-AGI-1 | 96.5% | 54.3% |
| CritPt | 30% | 0% |
| Chess Puzzles | 54% | 30% |
| LMArena Hard Prompts | 1468 | 1380 |
| Mystery Game Puzzles | 35% | 10% |
| DTBench | 93.3% | 80.5% |
| LMCA | 55% | 34.2% |
| Epoch Capabilities Index | 159.62 | 145.52 |
| SimpleBench | 48.9% | — |
| NYT Connections (extended) | 78.4% | — |
| EnigmaEval | — | 8.2% |
| Surface Evolver Bench | 83.8% | — |
| ForecastBench | — | 61 |
Math GPT-5.6 Terra leads
GPT-5.6 Terra: 81.6 (#12), GPT-5 Mini: 46.7 (#69)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 86% | 46.7% |
| FrontierMath Tier 4 | 70.7% | 12.2% |
| OTIS Mock AIME 2024-2025 | 99.7% | 86.7% |
| ProofBench | 74% | 9% |
| LMArena Math | 1466 | 1378 |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.6 Terra leads
GPT-5.6 Terra: 61.2 (#30), GPT-5 Mini: 45.6 (#86)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 93.3% | 75% |
| SimpleQA Verified | 43.2% | 21.6% |
| LMArena Expert | 1492 | 1379 |
| Humanity's Last Exam | — | 19.4% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| Vectara Hallucination Rate | — | 12.9% |
| GPQA (HELM) | — | 75.6% |
Multimodal GPT-5.6 Terra leads
GPT-5.6 Terra: 47.3 (#11), GPT-5 Mini: 35.6 (#85)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1271 | 1202 |
| VPCT | — | 40.2% |
| Blueprint-Bench 2 | 30.8% | — |
| Furniture Assembly | 54.2% | — |
| LMArena Document | 1472 | — |
Multilingual GPT-5.6 Terra leads
GPT-5.6 Terra: 54.4 (#44), GPT-5 Mini: 48.9 (#137)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1439 | 1363 |
| LMArena Chinese | 1513 | 1385 |
| LMArena French | 1471 | 1386 |
| LMArena German | 1460 | 1366 |
| LMArena Japanese | 1457 | 1341 |
| LMArena Korean | 1425 | 1308 |
| LMArena Russian | 1450 | 1362 |
| LMArena Spanish | 1448 | 1355 |
Instruction Following Too close to call
GPT-5.6 Terra: 76.4 (#40), GPT-5 Mini: 76.2 (#46)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1454 | 1357 |
| IFEval | — | 92.7% |
Long Context GPT-5.6 Terra leads
GPT-5.6 Terra: 44.4 (#68), GPT-5 Mini: 41.9 (#132)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1451 | 1355 |
| Fiction.LiveBench | — | 69.4% |
Writing & Preference GPT-5.6 Terra leads
GPT-5.6 Terra: 70.2 (#23), GPT-5 Mini: 55.2 (#148)
| Benchmark | GPT-5.6 Terra | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1447 | 1373 |
| LMArena Creative Writing | 1410 | 1325 |
| EQ-Bench Creative Writing | 1855 | 1313 |
| LMArena Multi-Turn | 1449 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
| EQ-Bench 4 | 1234 | — |
Frequently asked questions
Is GPT-5.6 Terra better than GPT-5 Mini?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 41.8 on the Noometry Index. GPT-5 Mini costs 6.5× less per token, which makes it the better buy when GPT-5.6 Terra's lead doesn't matter for your workload.
Which is cheaper, GPT-5.6 Terra 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-5.6 Terra lists at $2 and $12.
Is GPT-5.6 Terra or GPT-5 Mini better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 40.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 400K.
How many benchmarks do GPT-5.6 Terra and GPT-5 Mini share?
38 benchmarks have published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and GPT-5 Mini has 60.