Model comparison
GPT-6 Sol vs Grok 4.5
GPT-6 Sol is the stronger model overall, scoring 61.8 to 55.0 on the Noometry Index.
Last verified . 42 shared benchmarks.
Summary
- They share 42 benchmarks with published results for both. GPT-6 Sol scores higher in 6 categories and Grok 4.5 in 4 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 60.9.
- The biggest single-benchmark swing is FrontierMath Tier 4: 90% for GPT-6 Sol and 24.4% for Grok 4.5.
- Grok 4.5 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 4.5 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 61.8 | 55.0 |
| Released | 2026-09-22 | 2026-07-08 |
| 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 | 52 |
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 4.5: 52.2 (#35)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| DeepSWE | 68.8% | 53.8% |
| FrontierCode | 49.3% | 42.4% |
| LMArena WebDev | 1688 | 1553 |
| SciCode | 57.6% | 54.1% |
| LMArena Coding | 1447 | 1474 |
| ALE-Bench | 2,462 | 1,309 |
| WeirdML | — | 46.4% |
Agentic & Tool Use Grok 4.5 leads
GPT-6 Sol: 37.2 (#36), Grok 4.5: 44.4 (#17)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| APEX-Agents | 54.3% | 56.2% |
| GDP.pdf | 26.4% | 14% |
| Vending-Bench 2 | 14,428 | 3,887 |
| τ²-bench Banking | — | 47.9% |
| PostTrainBench | — | 23.4% |
| GBAEval | — | 65.4% |
| LMArena Search | — | 1213 |
Reasoning GPT-6 Sol leads
GPT-6 Sol: 74.0 (#9), Grok 4.5: 56.1 (#25)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| ARC-AGI-2 | 89.6% | 52.6% |
| NYT Connections (extended) | 90.1% | 79.9% |
| ARC-AGI-1 | 95.5% | 87.2% |
| CritPt | 30.9% | 15.4% |
| LMArena Hard Prompts | 1418 | 1462 |
| DTBench | 97.3% | 96.5% |
| LMCA | 59.1% | 45.2% |
| Epoch Capabilities Index | 162.72 | 153.92 |
| SimpleBench | — | 70% |
| Kagi LLM Benchmark | — | 83.5% |
| Chess Puzzles | — | 36% |
| EBR-Bench | 53.3% | — |
| Mystery Game Puzzles | 56% | — |
| Surface Evolver Bench | — | 74.4% |
Math GPT-6 Sol leads
GPT-6 Sol: 87.2 (#7), Grok 4.5: 60.9 (#35)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 89.8% | 57.2% |
| FrontierMath Tier 4 | 90% | 24.4% |
| OTIS Mock AIME 2024-2025 | 100% | 97.8% |
| ProofBench | 83% | 31% |
| LMArena Math | 1402 | 1459 |
Knowledge GPT-6 Sol leads
GPT-6 Sol: 64.8 (#15), Grok 4.5: 62.3 (#24)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 94.3% | 93.4% |
| SimpleQA Verified | 60.7% | 48.3% |
| LMArena Expert | 1439 | 1466 |
| Vectara Hallucination Rate | 6.5% | — |
Multimodal GPT-6 Sol leads
GPT-6 Sol: 47.6 (#10), Grok 4.5: 37.6 (#72)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| LMArena Vision | 1245 | 1288 |
| Blueprint-Bench 2 | 36.9% | 27.3% |
| Furniture Assembly | 58.3% | 22.5% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.5 leads
GPT-6 Sol: 50.5 (#118), Grok 4.5: 54.4 (#42)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1385 | 1440 |
| LMArena Chinese | 1405 | 1496 |
| LMArena French | 1410 | 1456 |
| LMArena German | 1390 | 1446 |
| LMArena Japanese | 1385 | 1428 |
| LMArena Korean | 1341 | 1404 |
| LMArena Russian | 1401 | 1448 |
| LMArena Spanish | 1384 | 1450 |
Instruction Following Grok 4.5 leads
GPT-6 Sol: 74.5 (#94), Grok 4.5: 76.0 (#48)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1412 | 1446 |
Long Context Grok 4.5 leads
GPT-6 Sol: 43.1 (#108), Grok 4.5: 44.8 (#56)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1411 | 1463 |
Writing & Preference GPT-6 Sol leads
GPT-6 Sol: 71.9 (#18), Grok 4.5: 65.8 (#42)
| Benchmark | GPT-6 Sol | Grok 4.5 |
|---|---|---|
| LMArena Text | 1395 | 1448 |
| LMArena Creative Writing | 1378 | 1442 |
| EQ-Bench Creative Writing | 2125 | 1579 |
| LMArena Multi-Turn | 1412 | 1456 |
Frequently asked questions
Is GPT-6 Sol better than Grok 4.5?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 55.0 on the Noometry Index.
Which is cheaper, GPT-6 Sol or Grok 4.5?
Grok 4.5 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 4.5 better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 52.2 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 4.5 share?
42 benchmarks have published results for both models. GPT-6 Sol has 45 scored results on Noometry and Grok 4.5 has 52.