Model comparison
GPT-5.2 vs Grok 4.3
GPT-5.2 is the stronger model overall, scoring 54.1 to 43.6 on the Noometry Index. Grok 4.3 costs 3.1× less per token, which makes it the better buy when GPT-5.2's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GPT-5.2 scores higher in 10 categories and Grok 4.3 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where GPT-5.2 leads 50.9 to 31.6.
- The biggest single-benchmark swing is NYT Connections (extended): 83.6% for GPT-5.2 and 55.2% for Grok 4.3.
- Grok 4.3 is cheaper at $1.25 / $2.50 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- Grok 4.3 accepts more context: 1M tokens versus 400K.
Side by side
| GPT-5.2 | Grok 4.3 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 54.1 | 43.6 |
| Released | 2025-12-11 | 2026-04-17 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1M |
| Max output | 128K | 30K |
| Input $ / M tokens | $1.75 | $1.25 |
| Output $ / M tokens | $14 | $2.50 |
| Results tracked | 67 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
GPT-5.2: 52.2 (#32), Grok 4.3: 41.3 (#117)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena WebDev | 1415 | 1357 |
| WeirdML | 72.2% | 49.9% |
| LMArena Coding | 1447 | 1415 |
| ALE-Bench | 1,294 | 944.17 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 72.8% | — |
| SWE-bench Multilingual | 66.7% | — |
| SciCode | — | 47.3% |
| GSO | 27.4% | — |
| AlgoTune | 2.05 | — |
Agentic & Tool Use GPT-5.2 leads
GPT-5.2: 40.0 (#24), Grok 4.3: 27.6 (#93)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Search | 1207 | 1165 |
| Vending-Bench 2 | 3,591 | 35.26 |
| 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% | — |
| GDP.pdf | — | 8% |
| METR Time Horizons | 75.3% | — |
Reasoning GPT-5.2 leads
GPT-5.2: 50.1 (#33), Grok 4.3: 34.9 (#60)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| NYT Connections (extended) | 83.6% | 55.2% |
| Chess Puzzles | 49% | 25% |
| LMArena Hard Prompts | 1445 | 1397 |
| DTBench | 90.9% | 90.7% |
| LMCA | 43.9% | 38.3% |
| Epoch Capabilities Index | 153.45 | 149.13 |
| ForecastBench | 60.1 | 60.3 |
| ARC-AGI-2 | 52.9% | — |
| SimpleBench | 45.8% | — |
| Kagi LLM Benchmark | 73.3% | — |
| ARC-AGI-1 | 86.2% | — |
| CritPt | — | 8% |
| EnigmaEval | 10.4% | — |
| EBR-Bench | 23% | — |
| Mystery Game Puzzles | 23% | — |
Math GPT-5.2 leads
GPT-5.2: 59.8 (#32), Grok 4.3: 45.8 (#60)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 67.4% | 42.8% |
| FrontierMath Tier 4 | 31.7% | 14.6% |
| OTIS Mock AIME 2024-2025 | 96.1% | 93.3% |
| ProofBench | 15% | 11% |
| LMArena Math | 1439 | 1388 |
| MathArena Final-Answer Competitions | 72% | — |
| FrontierMath (Feb 2025 set) | 40.7% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge GPT-5.2 leads
GPT-5.2: 59.2 (#29), Grok 4.3: 52.5 (#57)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| GPQA Diamond | 91.4% | 88.8% |
| SimpleQA Verified | 37.1% | 33.2% |
| LMArena Expert | 1445 | 1385 |
| Humanity's Last Exam | 27.8% | — |
| Vectara Hallucination Rate | 8.4% | — |
Multimodal GPT-5.2 leads
GPT-5.2: 50.9 (#7), Grok 4.3: 31.6 (#96)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Vision | 1268 | 1231 |
| VPCT | 84% | — |
| Blueprint-Bench 2 | — | 0% |
| Furniture Assembly | 38.3% | — |
| LMArena Document | 1405 | — |
Multilingual GPT-5.2 leads
GPT-5.2: 53.6 (#63), Grok 4.3: 50.7 (#114)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Non-English | 1425 | 1385 |
| LMArena Chinese | 1460 | 1423 |
| LMArena French | 1456 | 1412 |
| LMArena German | 1448 | 1395 |
| LMArena Japanese | 1421 | 1379 |
| LMArena Korean | 1390 | 1357 |
| LMArena Russian | 1440 | 1399 |
| LMArena Spanish | 1433 | 1402 |
Instruction Following GPT-5.2 leads
GPT-5.2: 74.9 (#83), Grok 4.3: 72.3 (#135)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Instruction Following | 1417 | 1366 |
Long Context GPT-5.2 leads
GPT-5.2: 44.0 (#74), Grok 4.3: 42.5 (#118)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Longer Query | 1428 | 1393 |
| CL-bench | 18.2% | — |
Writing & Preference GPT-5.2 leads
GPT-5.2: 67.0 (#30), Grok 4.3: 58.6 (#112)
| Benchmark | GPT-5.2 | Grok 4.3 |
|---|---|---|
| LMArena Text | 1439 | 1397 |
| LMArena Creative Writing | 1402 | 1380 |
| LMArena Multi-Turn | 1458 | 1406 |
| EQ-Bench Creative Writing | 1703 | — |
| EQ-Bench 4 | — | 1075 |
Frequently asked questions
Is GPT-5.2 better than Grok 4.3?
GPT-5.2 is the stronger model overall, scoring 54.1 to 43.6 on the Noometry Index. Grok 4.3 costs 3.1× less per token, which makes it the better buy when GPT-5.2's lead doesn't matter for your workload.
Which is cheaper, GPT-5.2 or Grok 4.3?
Grok 4.3 is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is GPT-5.2 or Grok 4.3 better for coding?
GPT-5.2 scores higher on coding benchmarks: 52.2 versus 41.3 in the Noometry coding category.
Which has the bigger context window?
Grok 4.3 does, with 1M tokens against 400K.
How many benchmarks do GPT-5.2 and Grok 4.3 share?
35 benchmarks have published results for both models. GPT-5.2 has 67 scored results on Noometry and Grok 4.3 has 40.