# GPT-5.2 vs Grok-2 (Dec 2024)

> GPT-5.2 is the stronger model overall, scoring 54.1 to 33.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-2-vs-grok-2
- Last updated: 2026-10-11
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-5.2 scores higher in 8 categories and Grok-2 (Dec 2024) in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.2 leads 60.0 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 96.1% for GPT-5.2 and 11.5% for Grok-2 (Dec 2024).

## Snapshot

| | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 54.1 | 33.7 |
| Rank | 34 | 239 |
| Context | 400K | — |
| Input $/M | $1.75 | — |
| Output $/M | $14 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.2: 51.6 (#37)
- Grok-2 (Dec 2024): 33.3 (#258)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 72.2% | 22.2% |
| LMArena Coding | 1447 | 1287 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| LMArena WebDev | 1416 | — |
| SWE-bench Multilingual | 66.7% | — |
| GSO | 27.4% | — |
| LiveBench Coding | — | 46.4% |
| ALE-Bench | 1,294 | — |
| AlgoTune | 2.05 | — |

## Agentic & Tool Use

- GPT-5.2: 40.2 (#24)
- Grok-2 (Dec 2024): —

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| Terminal-Bench | 64.9% | — |
| Berkeley Function Calling Leaderboard | 55.9% | — |
| GDPval | 49.7% | — |
| Remote Labor Index | 2.5% | — |
| τ²-bench Airline | 83% | — |
| τ²-bench Banking | 32.2% | — |
| τ²-bench Retail | 81.6% | — |
| τ²-bench Telecom | 89.7% | — |
| DeepResearch Bench | 41.1% | — |
| LMArena Search | 1207 | — |
| METR Time Horizons | 75.3% | — |
| Vending-Bench 2 | 3,591 | — |

## Reasoning

- GPT-5.2: 50.2 (#35)
- Grok-2 (Dec 2024): 16.9 (#299)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| SimpleBench | 45.8% | 22.7% |
| LMArena Hard Prompts | 1445 | 1272 |
| DTBench | 90.9% | 65.2% |
| Epoch Capabilities Index | 153.45 | 130.48 |
| ARC-AGI-2 | 52.9% | — |
| Kagi LLM Benchmark | 73.3% | — |
| NYT Connections (extended) | 83.6% | — |
| ARC-AGI-1 | 86.2% | — |
| Chess Puzzles | 49% | — |
| EnigmaEval | 10.4% | — |
| EBR-Bench | 23% | — |
| LiveBench Reasoning | — | 54.8% |
| Mystery Game Puzzles | 23% | — |
| LiveBench Data Analysis | — | 54.5% |
| LMCA | 43.9% | — |
| ForecastBench | 60.1 | — |
| LiveBench | — | 54.3% |

## Math

- GPT-5.2: 60.0 (#38)
- Grok-2 (Dec 2024): 20.8 (#284)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 11.5% |
| LMArena Math | 1440 | 1283 |
| FrontierMath (Feb 2025 set) | 40.7% | 0.7% |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 31.7% | — |
| MathArena Final-Answer Competitions | 72% | — |
| ProofBench | 15% | — |
| LiveBench Math | — | 54.9% |
| MATH Level 5 | — | 63.5% |
| FrontierMath Tier 4 (v1) | 18.8% | — |

## Knowledge

- GPT-5.2: 59.3 (#32)
- Grok-2 (Dec 2024): 29.8 (#233)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 91.4% | 53.8% |
| LMArena Expert | 1445 | 1254 |
| Humanity's Last Exam | 27.8% | — |
| SimpleQA Verified | 37.1% | — |
| Confabulations | — | 20.1% |
| Vectara Hallucination Rate | 8.4% | — |

## Multimodal

- GPT-5.2: 51.3 (#7)
- Grok-2 (Dec 2024): —

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Vision | 1268 | — |
| VPCT | 84% | — |
| Furniture Assembly | 38.3% | — |
| LMArena Document | 1405 | — |

## Multilingual

- GPT-5.2: 53.4 (#67)
- Grok-2 (Dec 2024): 43.1 (#188)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1425 | 1282 |
| LMArena Chinese | 1460 | 1289 |
| LMArena French | 1455 | 1318 |
| LMArena German | 1448 | 1287 |
| LMArena Japanese | 1420 | 1244 |
| LMArena Korean | 1392 | 1237 |
| LMArena Russian | 1440 | 1286 |
| LMArena Spanish | 1433 | 1281 |

## Instruction Following

- GPT-5.2: 74.7 (#89)
- Grok-2 (Dec 2024): 66.9 (#202)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Instruction Following | 1417 | 1270 |
| LiveBench Instruction Following | — | 69.6% |

## Long Context

- GPT-5.2: 44.0 (#78)
- Grok-2 (Dec 2024): 38.8 (#190)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1428 | 1276 |
| CL-bench | 18.2% | — |

## Writing & Preference

- GPT-5.2: 66.8 (#32)
- Grok-2 (Dec 2024): 48.6 (#198)

| Benchmark | GPT-5.2 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1439 | 1305 |
| LMArena Creative Writing | 1401 | 1284 |
| LMArena Multi-Turn | 1458 | 1290 |
| Short-Story Creative Writing | — | 63.6% |
| EQ-Bench Creative Writing | 1703 | — |
| LiveBench Language | — | 45.6% |

## FAQ

### Is GPT-5.2 better than Grok-2 (Dec 2024)?

GPT-5.2 is the stronger model overall, scoring 54.1 to 33.7 on the Noometry Index.

### Is GPT-5.2 or Grok-2 (Dec 2024) better for coding?

GPT-5.2 scores higher on coding benchmarks: 51.6 versus 33.3 in the Noometry coding category.

### How many benchmarks do GPT-5.2 and Grok-2 (Dec 2024) share?

24 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and Grok-2 (Dec 2024) has 34.