Model comparison
GPT-5.6 Terra vs Grok 4.7
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 53.1 on the Noometry Index. Grok 4.7 costs 1.5× 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 8 categories and Grok 4.7 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.6 to 57.8.
- The biggest single-benchmark swing is FrontierMath Tier 4: 70.7% for GPT-5.6 Terra and 17.1% for Grok 4.7.
- Grok 4.7 is cheaper at $2 / $6 per million input/output tokens, against $2 / $12 for GPT-5.6 Terra.
- GPT-5.6 Terra accepts more context: 1.05M tokens versus 500K.
Side by side
| GPT-5.6 Terra | Grok 4.7 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 59.2 | 53.1 |
| Released | 2026-07-09 | 2026-09-21 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $12 | $6 |
| Results tracked | 52 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5.6 Terra: 57.7 (#19), Grok 4.7: 58.0 (#18)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| FrontierCode | 41.3% | 47.6% |
| CursorBench | 41.3% | 46.3% |
| LMArena WebDev | 1522 | 1639 |
| SciCode | 55% | 57.8% |
| LMArena Coding | 1484 | 1427 |
| DeepSWE | 69.6% | — |
| FrontierSWE | — | 29.5% |
| WeirdML | 78.3% | — |
| ALE-Bench | 1,951 | — |
Agentic & Tool Use GPT-5.6 Terra leads
GPT-5.6 Terra: 40.1 (#25), Grok 4.7: 36.7 (#37)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| APEX-Agents | 58.2% | 54.6% |
| GDP.pdf | 24.7% | 22.8% |
| Vending-Bench 2 | 7,343 | 10,537 |
| BALROG | 53.2% | — |
Reasoning GPT-5.6 Terra leads
GPT-5.6 Terra: 60.7 (#21), Grok 4.7: 49.1 (#40)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| NYT Connections (extended) | 78.4% | 76.8% |
| CritPt | 30% | 18% |
| Chess Puzzles | 54% | 38% |
| LMArena Hard Prompts | 1468 | 1413 |
| Mystery Game Puzzles | 35% | 29% |
| DTBench | 93.3% | 96% |
| LMCA | 55% | 49.4% |
| Epoch Capabilities Index | 159.62 | 153.53 |
| ARC-AGI-2 | 83.9% | — |
| SimpleBench | 48.9% | — |
| Kagi LLM Benchmark | 51.3% | — |
| ARC-AGI-1 | 96.5% | — |
| Surface Evolver Bench | 83.8% | — |
Math GPT-5.6 Terra leads
GPT-5.6 Terra: 81.6 (#12), Grok 4.7: 57.8 (#39)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 86% | 53% |
| FrontierMath Tier 4 | 70.7% | 17.1% |
| OTIS Mock AIME 2024-2025 | 99.7% | 98.1% |
| ProofBench | 74% | 34% |
| LMArena Math | 1466 | 1407 |
Knowledge Grok 4.7 leads
GPT-5.6 Terra: 61.2 (#30), Grok 4.7: 62.8 (#22)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| GPQA Diamond | 93.3% | 92.7% |
| SimpleQA Verified | 43.2% | 56% |
| LMArena Expert | 1492 | 1422 |
Multimodal GPT-5.6 Terra leads
GPT-5.6 Terra: 47.3 (#11), Grok 4.7: 35.5 (#87)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| LMArena Vision | 1271 | 1228 |
| Blueprint-Bench 2 | 30.8% | 32.5% |
| Furniture Assembly | 54.2% | 20.8% |
| LMArena Document | 1472 | — |
Multilingual GPT-5.6 Terra leads
GPT-5.6 Terra: 54.4 (#44), Grok 4.7: 50.8 (#116)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| LMArena Non-English | 1439 | 1389 |
| LMArena Chinese | 1513 | 1455 |
| LMArena French | 1471 | 1455 |
| LMArena Russian | 1450 | 1397 |
| LMArena Spanish | 1448 | 1400 |
| LMArena German | 1460 | — |
| LMArena Japanese | 1457 | — |
| LMArena Korean | 1425 | — |
Instruction Following GPT-5.6 Terra leads
GPT-5.6 Terra: 76.4 (#40), Grok 4.7: 74.1 (#105)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| LMArena Instruction Following | 1454 | 1404 |
Long Context GPT-5.6 Terra leads
GPT-5.6 Terra: 44.4 (#68), Grok 4.7: 43.1 (#104)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| LMArena Longer Query | 1451 | 1413 |
Writing & Preference Too close to call
GPT-5.6 Terra: 70.2 (#23), Grok 4.7: 70.0 (#24)
| Benchmark | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| LMArena Text | 1447 | 1399 |
| LMArena Creative Writing | 1410 | 1391 |
| EQ-Bench Creative Writing | 1855 | 2007 |
| LMArena Multi-Turn | 1449 | 1393 |
| EQ-Bench 4 | 1234 | — |
Frequently asked questions
Is GPT-5.6 Terra better than Grok 4.7?
GPT-5.6 Terra is the stronger model overall, scoring 59.2 to 53.1 on the Noometry Index. Grok 4.7 costs 1.5× 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 Grok 4.7?
Grok 4.7 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-5.6 Terra lists at $2 and $12.
Is GPT-5.6 Terra or Grok 4.7 better for coding?
They score almost the same on coding (57.7 vs 58.0); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5.6 Terra does, with 1.05M tokens against 500K.
How many benchmarks do GPT-5.6 Terra and Grok 4.7 share?
38 benchmarks have published results for both models. GPT-5.6 Terra has 52 scored results on Noometry and Grok 4.7 has 39.