Model comparison
GPT-5.5 vs Grok 4.6
GPT-5.5 is the stronger model overall, scoring 63.4 to 56.9 on the Noometry Index. Grok 4.6 costs 3.8× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Last verified . 47 shared benchmarks.
Summary
- They share 47 benchmarks with published results for both. GPT-5.5 scores higher in 9 categories and Grok 4.6 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.7 to 67.0.
- The biggest single-benchmark swing is FrontierMath Tier 4: 72.5% for GPT-5.5 and 31.7% for Grok 4.6.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 500K.
Side by side
| GPT-5.5 | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 63.4 | 56.9 |
| Released | 2026-04-23 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $5 | $2 |
| Output $ / M tokens | $30 | $6 |
| Results tracked | 71 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5.5: 58.2 (#17), Grok 4.6: 58.5 (#16)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| DeepSWE | 67% | 67.5% |
| FrontierCode | 43% | 48% |
| LMArena WebDev | 1513 | 1617 |
| SciCode | 56.1% | 56.5% |
| WeirdML | 84.9% | 67.3% |
| LMArena Coding | 1494 | 1465 |
| ALE-Bench | 1,943 | 1,508 |
| SWE-bench Verified | 80.6% | — |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
| GSO | 40.2% | — |
| MirrorCode | 10% | — |
Agentic & Tool Use GPT-5.5 leads
GPT-5.5: 50.7 (#6), Grok 4.6: 39.4 (#27)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| APEX-Agents | 55.1% | 65.3% |
| GDP.pdf | 26% | 17.2% |
| Vending-Bench 2 | 7,524 | 9,047 |
| Terminal-Bench | 84.7% | — |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GBAEval | 53.2% | — |
| LMArena Search | 1242 | — |
Reasoning GPT-5.5 leads
GPT-5.5: 72.8 (#11), Grok 4.6: 61.4 (#20)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 85% | 67.1% |
| SimpleBench | 69% | 75.9% |
| NYT Connections (extended) | 96.2% | 80% |
| ARC-AGI-1 | 95% | 87.5% |
| CritPt | 27.1% | 19.7% |
| Chess Puzzles | 54% | 40% |
| EBR-Bench | 34.3% | 30.5% |
| LMArena Hard Prompts | 1489 | 1447 |
| Mystery Game Puzzles | 56% | 34% |
| DTBench | 96% | 97.3% |
| LMCA | 54.3% | 48.5% |
| Epoch Capabilities Index | 159.1 | 156.44 |
| Kagi LLM Benchmark | 88.8% | — |
| Surface Evolver Bench | 88.1% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 60.6 | — |
Math GPT-5.5 leads
GPT-5.5: 81.7 (#11), Grok 4.6: 67.0 (#24)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.3% | 66% |
| FrontierMath Tier 4 | 72.5% | 31.7% |
| OTIS Mock AIME 2024-2025 | 100% | 99.2% |
| ProofBench | 50% | 51% |
| LMArena Math | 1486 | 1423 |
| MathArena Final-Answer Competitions | 94.3% | — |
| FrontierMath (Feb 2025 set) | 51.7% | — |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | 35.4% | — |
Knowledge GPT-5.5 leads
GPT-5.5: 64.4 (#17), Grok 4.6: 63.3 (#20)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 94% | 94% |
| SimpleQA Verified | 63% | 49.3% |
| LMArena Expert | 1508 | 1467 |
| Vectara Hallucination Rate | 9.3% | — |
Multimodal GPT-5.5 leads
GPT-5.5: 46.9 (#12), Grok 4.6: 43.6 (#23)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1297 | 1263 |
| Blueprint-Bench 2 | 36.2% | 33.2% |
| Furniture Assembly | 44.2% | 40% |
| LMArena Document | 1486 | 1452 |
Multilingual GPT-5.5 leads
GPT-5.5: 56.4 (#20), Grok 4.6: 53.0 (#74)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1467 | 1420 |
| LMArena Chinese | 1533 | 1480 |
| LMArena French | 1486 | 1461 |
| LMArena German | 1480 | 1431 |
| LMArena Japanese | 1498 | 1376 |
| LMArena Korean | 1460 | 1397 |
| LMArena Russian | 1473 | 1422 |
| LMArena Spanish | 1468 | 1404 |
Instruction Following GPT-5.5 leads
GPT-5.5: 77.5 (#18), Grok 4.6: 75.4 (#63)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1479 | 1431 |
Long Context GPT-5.5 leads
GPT-5.5: 48.3 (#12), Grok 4.6: 44.5 (#66)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1484 | 1454 |
| CL-bench Life | 22.2% | — |
Writing & Preference GPT-5.5 leads
GPT-5.5: 72.7 (#13), Grok 4.6: 62.3 (#80)
| Benchmark | GPT-5.5 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1472 | 1428 |
| LMArena Creative Writing | 1455 | 1428 |
| LMArena Multi-Turn | 1476 | 1425 |
| EQ-Bench Creative Writing | 1844 | — |
| EQ-Bench 4 | 1315 | — |
Frequently asked questions
Is GPT-5.5 better than Grok 4.6?
GPT-5.5 is the stronger model overall, scoring 63.4 to 56.9 on the Noometry Index. Grok 4.6 costs 3.8× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.
Which is cheaper, GPT-5.5 or Grok 4.6?
Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-5.5 lists at $5 and $30.
Is GPT-5.5 or Grok 4.6 better for coding?
They score almost the same on coding (58.2 vs 58.5); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 500K.
How many benchmarks do GPT-5.5 and Grok 4.6 share?
47 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Grok 4.6 has 49.