Model comparison
GPT-5.2 vs GPT-5.6 Terra
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 54.1 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. GPT-5.2 scores higher in 2 categories and GPT-5.6 Terra in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 60.0.
- The biggest single-benchmark swing is ProofBench: 15% for GPT-5.2 and 74% for GPT-5.6 Terra.
- GPT-5.6 Terra is cheaper at $2 / $12 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-5.2 | GPT-5.6 Terra | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 54.1 | 59.2 |
| Released | 2025-12-11 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $1.75 | $2 |
| Output $ / M tokens | $14 | $12 |
| Results tracked | 67 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
GPT-5.2: 51.6 (#37), GPT-5.6 Terra: 57.7 (#19)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena WebDev | 1416 | 1522 |
| WeirdML | 72.2% | 78.3% |
| LMArena Coding | 1447 | 1484 |
| ALE-Bench | 1,294 | 1,951 |
| SWE-bench Verified | 73.8% | — |
| DeepSWE | — | 69.6% |
| FrontierCode | — | 41.3% |
| SWE-bench Verified (bash only) | 72.8% | — |
| CursorBench | — | 41.3% |
| SWE-bench Multilingual | 66.7% | — |
| SciCode | — | 55% |
| GSO | 27.4% | — |
| AlgoTune | 2.05 | — |
Agentic & Tool Use Too close to call
GPT-5.2: 40.2 (#24), GPT-5.6 Terra: 40.1 (#25)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| Vending-Bench 2 | 3,591 | 7,343 |
| Terminal-Bench | 64.9% | — |
| APEX-Agents | — | 58.2% |
| Berkeley Function Calling Leaderboard | 55.9% | — |
| GDPval | 49.7% | — |
| Remote Labor Index | 2.5% | — |
| τ²-bench Airline | 83% | — |
| τ²-bench Banking | 32.2% | — |
| τ²-bench Retail | 81.6% | — |
| τ²-bench Telecom | 89.7% | — |
| DeepResearch Bench | 41.1% | — |
| BALROG | — | 53.2% |
| GDP.pdf | — | 24.7% |
| LMArena Search | 1207 | — |
| METR Time Horizons | 75.3% | — |
Reasoning GPT-5.6 Terra leads
GPT-5.2: 50.2 (#35), GPT-5.6 Terra: 60.7 (#21)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| ARC-AGI-2 | 52.9% | 83.9% |
| SimpleBench | 45.8% | 48.9% |
| Kagi LLM Benchmark | 73.3% | 51.3% |
| NYT Connections (extended) | 83.6% | 78.4% |
| ARC-AGI-1 | 86.2% | 96.5% |
| Chess Puzzles | 49% | 54% |
| LMArena Hard Prompts | 1445 | 1468 |
| Mystery Game Puzzles | 23% | 35% |
| DTBench | 90.9% | 93.3% |
| LMCA | 43.9% | 55% |
| Epoch Capabilities Index | 153.45 | 159.62 |
| CritPt | — | 30% |
| EnigmaEval | 10.4% | — |
| EBR-Bench | 23% | — |
| Surface Evolver Bench | — | 83.8% |
| ForecastBench | 60.1 | — |
Math GPT-5.6 Terra leads
GPT-5.2: 60.0 (#38), GPT-5.6 Terra: 81.6 (#12)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 86% |
| FrontierMath Tier 4 | 31.7% | 70.7% |
| OTIS Mock AIME 2024-2025 | 96.1% | 99.7% |
| ProofBench | 15% | 74% |
| LMArena Math | 1440 | 1466 |
| MathArena Final-Answer Competitions | 72% | — |
| FrontierMath (Feb 2025 set) | 40.7% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge GPT-5.6 Terra leads
GPT-5.2: 59.3 (#32), GPT-5.6 Terra: 61.2 (#30)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| GPQA Diamond | 91.4% | 93.3% |
| SimpleQA Verified | 37.1% | 43.2% |
| LMArena Expert | 1445 | 1492 |
| Humanity's Last Exam | 27.8% | — |
| Vectara Hallucination Rate | 8.4% | — |
Multimodal GPT-5.2 leads
GPT-5.2: 51.3 (#7), GPT-5.6 Terra: 47.3 (#11)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | 1268 | 1271 |
| Furniture Assembly | 38.3% | 54.2% |
| LMArena Document | 1405 | 1472 |
| VPCT | 84% | — |
| Blueprint-Bench 2 | — | 30.8% |
Multilingual Too close to call
GPT-5.2: 53.4 (#67), GPT-5.6 Terra: 54.4 (#44)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1425 | 1439 |
| LMArena Chinese | 1460 | 1513 |
| LMArena French | 1455 | 1471 |
| LMArena German | 1448 | 1460 |
| LMArena Japanese | 1420 | 1457 |
| LMArena Korean | 1392 | 1425 |
| LMArena Russian | 1440 | 1450 |
| LMArena Spanish | 1433 | 1448 |
Instruction Following GPT-5.6 Terra leads
GPT-5.2: 74.7 (#89), GPT-5.6 Terra: 76.4 (#40)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1417 | 1454 |
Long Context Too close to call
GPT-5.2: 44.0 (#78), GPT-5.6 Terra: 44.4 (#68)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1428 | 1451 |
| CL-bench | 18.2% | — |
Writing & Preference GPT-5.6 Terra leads
GPT-5.2: 66.8 (#32), GPT-5.6 Terra: 70.2 (#23)
| Benchmark | GPT-5.2 | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1439 | 1447 |
| LMArena Creative Writing | 1401 | 1410 |
| EQ-Bench Creative Writing | 1703 | 1855 |
| LMArena Multi-Turn | 1458 | 1449 |
| EQ-Bench 4 | — | 1234 |
Frequently asked questions
Is GPT-5.2 better than GPT-5.6 Terra?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 54.1 on the Noometry Index.
Which is cheaper, GPT-5.2 or GPT-5.6 Terra?
GPT-5.6 Terra is cheaper. It lists at $2 per million input tokens and $12 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is GPT-5.2 or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.7 versus 51.6 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-5.2 and GPT-5.6 Terra share?
41 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and GPT-5.6 Terra has 52.