Model comparison
GPT-5.6 Terra vs GPT-6 Sol
GPT-5.6 Terra is the stronger model overall, scoring 59.1 to 49.2 on the Noometry Index.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. GPT-5.6 Terra scores higher in 7 categories and GPT-6 Sol in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.5 to 35.8.
- The biggest single-benchmark swing is NYT Connections (extended): 78.4% for GPT-5.6 Terra and 90.1% for GPT-6 Sol.
- GPT-6 Sol is cheaper at $2 / $10 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
Side by side
| GPT-5.6 Terra | GPT-6 Sol | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 59.1 | 49.2 |
| Released | 2026-07-09 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $12 | $10 |
| Results tracked | 52 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-5.6 Terra: 57.1 (#16), GPT-6 Sol: 59.6 (#11)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| DeepSWE | 69.6% | 68.8% |
| FrontierCode | 41.3% | 49.3% |
| LMArena WebDev | 1520 | 1692 |
| LMArena Coding | 1484 | 1453 |
| CursorBench | 41.3% | — |
| SciCode | 55% | — |
| WeirdML | 78.3% | — |
| ALE-Bench | 1,951 | — |
Agentic & Tool Use GPT-5.6 Terra leads
GPT-5.6 Terra: 40.2 (#23), GPT-6 Sol: 33.5 (#49)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| GDP.pdf | 24.7% | 28% |
| APEX-Agents | 58.2% | — |
| BALROG | 53.2% | — |
| Vending-Bench 2 | 7,343 | — |
Reasoning GPT-5.6 Terra leads
GPT-5.6 Terra: 60.6 (#19), GPT-6 Sol: 49.3 (#34)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| NYT Connections (extended) | 78.4% | 90.1% |
| LMArena Hard Prompts | 1467 | 1420 |
| ARC-AGI-2 | 83.9% | — |
| SimpleBench | 48.9% | — |
| Kagi LLM Benchmark | 51.3% | — |
| ARC-AGI-1 | 96.5% | — |
| CritPt | 30% | — |
| Chess Puzzles | 54% | — |
| EBR-Bench | — | 53.3% |
| Mystery Game Puzzles | 35% | — |
| DTBench | 93.3% | — |
| LMCA | 55% | — |
| Surface Evolver Bench | 83.8% | — |
| Epoch Capabilities Index | 159.31 | — |
Math GPT-5.6 Terra leads
GPT-5.6 Terra: 81.5 (#10), GPT-6 Sol: 35.8 (#135)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Math | 1463 | 1404 |
| FrontierMath (Tiers 1-3) | 86% | — |
| FrontierMath Tier 4 | 70.7% | — |
| OTIS Mock AIME 2024-2025 | 99.7% | — |
| ProofBench | 74% | — |
Knowledge GPT-5.6 Terra leads
GPT-5.6 Terra: 61.3 (#27), GPT-6 Sol: 43.0 (#87)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Expert | 1492 | 1449 |
| GPQA Diamond | 93.3% | — |
| SimpleQA Verified | 43.2% | — |
| Vectara Hallucination Rate | — | 6.5% |
Multimodal Too close to call
GPT-5.6 Terra: 47.5 (#10), GPT-6 Sol: 47.6 (#9)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| Furniture Assembly | 54.2% | 58.3% |
| LMArena Vision | 1271 | — |
| Blueprint-Bench 2 | 30.8% | — |
| LMArena Document | 1472 | — |
Multilingual GPT-5.6 Terra leads
GPT-5.6 Terra: 54.5 (#42), GPT-6 Sol: 51.0 (#113)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1439 | 1389 |
| LMArena Chinese | 1514 | 1416 |
| LMArena Russian | 1450 | 1407 |
| LMArena Spanish | 1455 | 1377 |
| LMArena French | 1476 | — |
| LMArena German | 1462 | — |
| LMArena Japanese | 1446 | — |
| LMArena Korean | 1420 | — |
Instruction Following GPT-5.6 Terra leads
GPT-5.6 Terra: 76.6 (#38), GPT-6 Sol: 74.4 (#97)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1453 | 1406 |
Long Context GPT-5.6 Terra leads
GPT-5.6 Terra: 44.4 (#64), GPT-6 Sol: 43.2 (#100)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1452 | 1415 |
Writing & Preference GPT-6 Sol leads
GPT-5.6 Terra: 70.4 (#21), GPT-6 Sol: 72.3 (#15)
| Benchmark | GPT-5.6 Terra | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1446 | 1396 |
| LMArena Creative Writing | 1411 | 1392 |
| EQ-Bench Creative Writing | 1855 | 2125 |
| LMArena Multi-Turn | 1451 | 1405 |
| EQ-Bench 4 | 1234 | — |
Frequently asked questions
Is GPT-5.6 Terra better than GPT-6 Sol?
GPT-5.6 Terra is the stronger model overall, scoring 59.1 to 49.2 on the Noometry Index.
Which is cheaper, GPT-5.6 Terra or GPT-6 Sol?
GPT-6 Sol is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Is GPT-5.6 Terra or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 59.6 versus 57.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do GPT-5.6 Terra and GPT-6 Sol share?
20 benchmarks have published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and GPT-6 Sol has 22.