Model comparison
GPT-5.2 vs Grok 4
GPT-5.2 is the stronger model overall, scoring 54.1 to 48.1 on the Noometry Index.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-5.2 scores higher in 8 categories and Grok 4 in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 44.0.
- The biggest single-benchmark swing is Terminal-Bench: 64.9% for GPT-5.2 and 27.2% for Grok 4.
Side by side
| GPT-5.2 | Grok 4 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 54.1 | 48.1 |
| Released | 2025-12-11 | 2025-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.75 | — |
| Output $ / M tokens | $14 | — |
| Results tracked | 67 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
GPT-5.2: 51.6 (#37), Grok 4: 50.3 (#46)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| WeirdML | 72.2% | 45.7% |
| LMArena Coding | 1447 | 1408 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1416 | — |
| SWE-bench Multilingual | 66.7% | — |
| GSO | 27.4% | — |
| ALE-Bench | 1,294 | — |
| AlgoTune | 2.05 | — |
Agentic & Tool Use GPT-5.2 leads
GPT-5.2: 40.2 (#24), Grok 4: 32.3 (#68)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| Terminal-Bench | 64.9% | 27.2% |
| Berkeley Function Calling Leaderboard | 55.9% | 63% |
| GDPval | 49.7% | 21.1% |
| DeepResearch Bench | 41.1% | 47.3% |
| LMArena Search | 1207 | 1142 |
| METR Time Horizons | 75.3% | 66.6% |
| Remote Labor Index | 2.5% | — |
| τ²-bench Airline | 83% | — |
| τ²-bench Banking | 32.2% | — |
| τ²-bench Retail | 81.6% | — |
| τ²-bench Telecom | 89.7% | — |
| Cybench | — | 43% |
| BALROG | — | 43.6% |
| Vending-Bench 2 | 3,591 | — |
Reasoning GPT-5.2 leads
GPT-5.2: 50.2 (#35), Grok 4: 36.7 (#65)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| ARC-AGI-2 | 52.9% | 16% |
| SimpleBench | 45.8% | 60.5% |
| Kagi LLM Benchmark | 73.3% | 73.6% |
| ARC-AGI-1 | 86.2% | 66.7% |
| Chess Puzzles | 49% | 28% |
| LMArena Hard Prompts | 1445 | 1409 |
| Epoch Capabilities Index | 153.45 | 146.44 |
| ForecastBench | 60.1 | 60.9 |
| NYT Connections (extended) | 83.6% | — |
| EnigmaEval | 10.4% | — |
| EBR-Bench | 23% | — |
| Mystery Game Puzzles | 23% | — |
| DTBench | 90.9% | — |
| LMCA | 43.9% | — |
Math GPT-5.2 leads
GPT-5.2: 60.0 (#38), Grok 4: 48.4 (#64)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 84% |
| LMArena Math | 1440 | 1422 |
| FrontierMath (Feb 2025 set) | 40.7% | 19.7% |
| FrontierMath Tier 4 (v1) | 18.8% | 2.1% |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 31.7% | — |
| MathArena Final-Answer Competitions | 72% | — |
| ProofBench | 15% | — |
| Omni-MATH | — | 60.3% |
Knowledge GPT-5.2 leads
GPT-5.2: 59.3 (#32), Grok 4: 53.8 (#55)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| GPQA Diamond | 91.4% | 87% |
| LMArena Expert | 1445 | 1415 |
| Humanity's Last Exam | 27.8% | — |
| SimpleQA Verified | 37.1% | — |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| Vectara Hallucination Rate | 8.4% | — |
| GPQA (HELM) | — | 72.7% |
Multimodal GPT-5.2 leads
GPT-5.2: 51.3 (#7), Grok 4: 33.7 (#94)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| LMArena Vision | 1268 | 1210 |
| GeoBench | — | 45% |
| VPCT | 84% | — |
| Furniture Assembly | 38.3% | — |
| LMArena Document | 1405 | — |
Multilingual GPT-5.2 leads
GPT-5.2: 53.4 (#67), Grok 4: 51.8 (#103)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| LMArena Non-English | 1425 | 1403 |
| LMArena Chinese | 1460 | 1427 |
| LMArena French | 1455 | 1418 |
| LMArena German | 1448 | 1429 |
| LMArena Japanese | 1420 | 1394 |
| LMArena Korean | 1392 | 1377 |
| LMArena Russian | 1440 | 1410 |
| LMArena Spanish | 1433 | 1420 |
Instruction Following Grok 4 leads
GPT-5.2: 74.7 (#89), Grok 4: 79.2 (#5)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1417 | 1387 |
| IFEval | — | 94.9% |
Long Context Grok 4 leads
GPT-5.2: 44.0 (#78), Grok 4: 63.1 (#4)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1428 | 1409 |
| Fiction.LiveBench | — | 94.4% |
| CL-bench | 18.2% | — |
Writing & Preference GPT-5.2 leads
GPT-5.2: 66.8 (#32), Grok 4: 58.5 (#116)
| Benchmark | GPT-5.2 | Grok 4 |
|---|---|---|
| LMArena Text | 1439 | 1411 |
| LMArena Creative Writing | 1401 | 1397 |
| LMArena Multi-Turn | 1458 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 1703 | — |
| WildBench | — | 79.7% |
Frequently asked questions
Is GPT-5.2 better than Grok 4?
GPT-5.2 is the stronger model overall, scoring 54.1 to 48.1 on the Noometry Index.
Is GPT-5.2 or Grok 4 better for coding?
GPT-5.2 scores higher on coding benchmarks: 51.6 versus 50.3 in the Noometry coding category.
How many benchmarks do GPT-5.2 and Grok 4 share?
36 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and Grok 4 has 48.