Model comparison
GPT-6 Sol vs Grok 4.6
GPT-6 Sol is the stronger model overall, scoring 61.8 to 56.9 on the Noometry Index.
Last verified . 43 shared benchmarks.
Summary
- They share 43 benchmarks with published results for both. GPT-6 Sol scores higher in 6 categories and Grok 4.6 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 67.0.
- The biggest single-benchmark swing is FrontierMath Tier 4: 90% for GPT-6 Sol and 31.7% for Grok 4.6.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 500K.
Side by side
| GPT-6 Sol | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 61.8 | 56.9 |
| Released | 2026-09-22 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $10 | $6 |
| Results tracked | 45 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-6 Sol: 60.1 (#11), Grok 4.6: 58.5 (#16)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| DeepSWE | 68.8% | 67.5% |
| FrontierCode | 49.3% | 48% |
| LMArena WebDev | 1688 | 1617 |
| SciCode | 57.6% | 56.5% |
| LMArena Coding | 1447 | 1465 |
| ALE-Bench | 2,462 | 1,508 |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
| WeirdML | — | 67.3% |
Agentic & Tool Use Grok 4.6 leads
GPT-6 Sol: 37.2 (#36), Grok 4.6: 39.4 (#27)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| APEX-Agents | 54.3% | 65.3% |
| GDP.pdf | 26.4% | 17.2% |
| Vending-Bench 2 | 14,428 | 9,047 |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), Grok 4.6: 61.4 (#20)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 89.6% | 67.1% |
| NYT Connections (extended) | 90.1% | 80% |
| ARC-AGI-1 | 95.5% | 87.5% |
| CritPt | 30.9% | 19.7% |
| EBR-Bench | 53.3% | 30.5% |
| LMArena Hard Prompts | 1418 | 1447 |
| Mystery Game Puzzles | 56% | 34% |
| DTBench | 97.3% | 97.3% |
| LMCA | 59.1% | 48.5% |
| Epoch Capabilities Index | 162.72 | 156.44 |
| SimpleBench | — | 75.9% |
| Chess Puzzles | — | 40% |
Math GPT-6 Sol leads
GPT-6 Sol: 87.2 (#7), Grok 4.6: 67.0 (#24)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 89.8% | 66% |
| FrontierMath Tier 4 | 90% | 31.7% |
| OTIS Mock AIME 2024-2025 | 100% | 99.2% |
| ProofBench | 83% | 51% |
| LMArena Math | 1402 | 1423 |
Knowledge GPT-6 Sol leads
GPT-6 Sol: 64.8 (#15), Grok 4.6: 63.3 (#20)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 94.3% | 94% |
| SimpleQA Verified | 60.7% | 49.3% |
| LMArena Expert | 1439 | 1467 |
| Vectara Hallucination Rate | 6.5% | — |
Multimodal GPT-6 Sol leads
GPT-6 Sol: 47.6 (#10), Grok 4.6: 43.6 (#23)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1245 | 1263 |
| Blueprint-Bench 2 | 36.9% | 33.2% |
| Furniture Assembly | 58.3% | 40% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.6 leads
GPT-6 Sol: 50.5 (#118), Grok 4.6: 53.0 (#74)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1385 | 1420 |
| LMArena Chinese | 1405 | 1480 |
| LMArena French | 1410 | 1461 |
| LMArena German | 1390 | 1431 |
| LMArena Japanese | 1385 | 1376 |
| LMArena Korean | 1341 | 1397 |
| LMArena Russian | 1401 | 1422 |
| LMArena Spanish | 1384 | 1404 |
Instruction Following Too close to call
GPT-6 Sol: 74.5 (#94), Grok 4.6: 75.4 (#63)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1412 | 1431 |
Long Context Grok 4.6 leads
GPT-6 Sol: 43.1 (#108), Grok 4.6: 44.5 (#66)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1411 | 1454 |
Writing & Preference GPT-6 Sol leads
GPT-6 Sol: 71.9 (#18), Grok 4.6: 62.3 (#80)
| Benchmark | GPT-6 Sol | Grok 4.6 |
|---|---|---|
| LMArena Text | 1395 | 1428 |
| LMArena Creative Writing | 1378 | 1428 |
| LMArena Multi-Turn | 1412 | 1425 |
| EQ-Bench Creative Writing | 2125 | — |
Frequently asked questions
Is GPT-6 Sol better than Grok 4.6?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 56.9 on the Noometry Index.
Which is cheaper, GPT-6 Sol or Grok 4.6?
Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-6 Sol or Grok 4.6 better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 58.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 500K.
How many benchmarks do GPT-6 Sol and Grok 4.6 share?
43 benchmarks have published results for both models. GPT-6 Sol has 45 scored results on Noometry and Grok 4.6 has 49.