Model comparison
GPT-6 Luna vs Grok 4
Grok 4 is the stronger model overall, scoring 47.8 to 45.1 on the Noometry Index.
Last verified . 13 shared benchmarks.
Summary
- They share 13 benchmarks with published results for both. GPT-6 Luna scores higher in 2 categories and Grok 4 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 43.3.
Side by side
| GPT-6 Luna | Grok 4 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 45.1 | 47.8 |
| Released | 2026-09-22 | 2025-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.10 | — |
| Output $ / M tokens | $0.50 | — |
| Results tracked | 18 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 54.7 (#24), Grok 4: 49.9 (#44)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Coding | 1448 | 1408 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1583 | — |
| WeirdML | — | 45.7% |
Agentic & Tool Use Not comparable
GPT-6 Luna: —, Grok 4: 32.1 (#62)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 21.1% |
| Cybench | — | 43% |
| DeepResearch Bench | — | 47.3% |
| BALROG | — | 43.6% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 66.6% |
Reasoning Grok 4 leads
GPT-6 Luna: 31.7 (#68), Grok 4: 36.9 (#59)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Hard Prompts | 1415 | 1409 |
| ARC-AGI-2 | — | 16% |
| SimpleBench | — | 60.5% |
| Kagi LLM Benchmark | — | 73.6% |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | — | 66.7% |
| Chess Puzzles | — | 28% |
| Epoch Capabilities Index | — | 146.46 |
| ForecastBench | — | 60.9 |
Math Grok 4 leads
GPT-6 Luna: 35.6 (#140), Grok 4: 46.5 (#58)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Math | 1396 | 1422 |
| OTIS Mock AIME 2024-2025 | — | 84% |
| Omni-MATH | — | 60.3% |
| FrontierMath (Feb 2025 set) | — | 19.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Grok 4 leads
GPT-6 Luna: 40.1 (#110), Grok 4: 53.7 (#50)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Expert | 1458 | 1417 |
| GPQA Diamond | — | 87% |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| GPQA (HELM) | — | 72.7% |
Multimodal GPT-6 Luna leads
GPT-6 Luna: 42.0 (#27), Grok 4: 33.1 (#89)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Vision | — | 1211 |
| GeoBench | — | 45% |
| Furniture Assembly | 44.2% | — |
Multilingual Too close to call
GPT-6 Luna: 51.0 (#111), Grok 4: 52.0 (#98)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Non-English | 1390 | 1403 |
| LMArena Chinese | 1425 | 1428 |
| LMArena Russian | 1392 | 1409 |
| LMArena Spanish | 1416 | 1419 |
| LMArena French | — | 1418 |
| LMArena German | — | 1429 |
| LMArena Japanese | — | 1396 |
| LMArena Korean | — | 1376 |
Instruction Following Grok 4 leads
GPT-6 Luna: 74.7 (#85), Grok 4: 79.4 (#4)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1414 | 1387 |
| IFEval | — | 94.9% |
Long Context Grok 4 leads
GPT-6 Luna: 43.3 (#95), Grok 4: 63.1 (#4)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1417 | 1409 |
| Fiction.LiveBench | — | 94.4% |
Writing & Preference Too close to call
GPT-6 Luna: 58.4 (#115), Grok 4: 58.7 (#111)
| Benchmark | GPT-6 Luna | Grok 4 |
|---|---|---|
| LMArena Text | 1392 | 1412 |
| LMArena Creative Writing | 1357 | 1397 |
| LMArena Multi-Turn | 1400 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| WildBench | — | 79.7% |
Frequently asked questions
Is GPT-6 Luna better than Grok 4?
Grok 4 is the stronger model overall, scoring 47.8 to 45.1 on the Noometry Index.
Is GPT-6 Luna or Grok 4 better for coding?
GPT-6 Luna scores higher on coding benchmarks: 54.7 versus 49.9 in the Noometry coding category.
How many benchmarks do GPT-6 Luna and Grok 4 share?
13 benchmarks have published results for both models. GPT-6 Luna has 18 scored results on Noometry and Grok 4 has 48.