Model comparison
GPT-5.5 vs Grok 4.5
GPT-5.5 is the stronger model overall, scoring 63.4 to 55.0 on the Noometry Index. Grok 4.5 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 . 52 shared benchmarks.
Summary
- They share 52 benchmarks with published results for both. GPT-5.5 scores higher in 10 categories and Grok 4.5 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.7 to 60.9.
- The biggest single-benchmark swing is FrontierMath Tier 4: 72.5% for GPT-5.5 and 24.4% for Grok 4.5.
- Grok 4.5 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 4.5 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 63.4 | 55.0 |
| Released | 2026-04-23 | 2026-07-08 |
| 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 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.5 leads
GPT-5.5: 58.2 (#17), Grok 4.5: 52.2 (#35)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| DeepSWE | 67% | 53.8% |
| FrontierCode | 43% | 42.4% |
| LMArena WebDev | 1513 | 1553 |
| SciCode | 56.1% | 54.1% |
| WeirdML | 84.9% | 46.4% |
| LMArena Coding | 1494 | 1474 |
| ALE-Bench | 1,943 | 1,309 |
| SWE-bench Verified | 80.6% | — |
| GSO | 40.2% | — |
| MirrorCode | 10% | — |
Agentic & Tool Use GPT-5.5 leads
GPT-5.5: 50.7 (#6), Grok 4.5: 44.4 (#17)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| APEX-Agents | 55.1% | 56.2% |
| τ²-bench Banking | 44.6% | 47.9% |
| PostTrainBench | 27.2% | 23.4% |
| GBAEval | 53.2% | 65.4% |
| GDP.pdf | 26% | 14% |
| LMArena Search | 1242 | 1213 |
| Vending-Bench 2 | 7,524 | 3,887 |
| Terminal-Bench | 84.7% | — |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| DeepResearch Bench | 54% | — |
| ExploitBench | 47.4% | — |
Reasoning GPT-5.5 leads
GPT-5.5: 72.8 (#11), Grok 4.5: 56.1 (#25)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| ARC-AGI-2 | 85% | 52.6% |
| SimpleBench | 69% | 70% |
| Kagi LLM Benchmark | 88.8% | 83.5% |
| NYT Connections (extended) | 96.2% | 79.9% |
| ARC-AGI-1 | 95% | 87.2% |
| CritPt | 27.1% | 15.4% |
| Chess Puzzles | 54% | 36% |
| LMArena Hard Prompts | 1489 | 1462 |
| DTBench | 96% | 96.5% |
| LMCA | 54.3% | 45.2% |
| Surface Evolver Bench | 88.1% | 74.4% |
| Epoch Capabilities Index | 159.1 | 153.92 |
| EBR-Bench | 34.3% | — |
| Mystery Game Puzzles | 56% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 60.6 | — |
Math GPT-5.5 leads
GPT-5.5: 81.7 (#11), Grok 4.5: 60.9 (#35)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.3% | 57.2% |
| FrontierMath Tier 4 | 72.5% | 24.4% |
| OTIS Mock AIME 2024-2025 | 100% | 97.8% |
| ProofBench | 50% | 31% |
| LMArena Math | 1486 | 1459 |
| 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 4.5: 62.3 (#24)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 94% | 93.4% |
| SimpleQA Verified | 63% | 48.3% |
| LMArena Expert | 1508 | 1466 |
| Vectara Hallucination Rate | 9.3% | — |
Multimodal GPT-5.5 leads
GPT-5.5: 46.9 (#12), Grok 4.5: 37.6 (#72)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| LMArena Vision | 1297 | 1288 |
| Blueprint-Bench 2 | 36.2% | 27.3% |
| Furniture Assembly | 44.2% | 22.5% |
| LMArena Document | 1486 | 1452 |
Multilingual GPT-5.5 leads
GPT-5.5: 56.4 (#20), Grok 4.5: 54.4 (#42)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1467 | 1440 |
| LMArena Chinese | 1533 | 1496 |
| LMArena French | 1486 | 1456 |
| LMArena German | 1480 | 1446 |
| LMArena Japanese | 1498 | 1428 |
| LMArena Korean | 1460 | 1404 |
| LMArena Russian | 1473 | 1448 |
| LMArena Spanish | 1468 | 1450 |
Instruction Following GPT-5.5 leads
GPT-5.5: 77.5 (#18), Grok 4.5: 76.0 (#48)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1479 | 1446 |
Long Context GPT-5.5 leads
GPT-5.5: 48.3 (#12), Grok 4.5: 44.8 (#56)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1484 | 1463 |
| CL-bench Life | 22.2% | — |
Writing & Preference GPT-5.5 leads
GPT-5.5: 72.7 (#13), Grok 4.5: 65.8 (#42)
| Benchmark | GPT-5.5 | Grok 4.5 |
|---|---|---|
| LMArena Text | 1472 | 1448 |
| LMArena Creative Writing | 1455 | 1442 |
| EQ-Bench Creative Writing | 1844 | 1579 |
| LMArena Multi-Turn | 1476 | 1456 |
| EQ-Bench 4 | 1315 | — |
Frequently asked questions
Is GPT-5.5 better than Grok 4.5?
GPT-5.5 is the stronger model overall, scoring 63.4 to 55.0 on the Noometry Index. Grok 4.5 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 4.5?
Grok 4.5 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 4.5 better for coding?
GPT-5.5 scores higher on coding benchmarks: 58.2 versus 52.2 in the Noometry coding category.
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 4.5 share?
52 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and Grok 4.5 has 52.