Model comparison
GPT-6 Luna vs Grok 4.6
Grok 4.6 is the stronger model overall, scoring 56.9 to 53.3 on the Noometry Index. GPT-6 Luna costs 15× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Last verified . 42 shared benchmarks.
Summary
- They share 42 benchmarks with published results for both. GPT-6 Luna scores higher in 1 category and Grok 4.6 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 48.2.
- The biggest single-benchmark swing is Mystery Game Puzzles: 7% for GPT-6 Luna and 34% for Grok 4.6.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $2 / $6 for Grok 4.6.
- GPT-6 Luna accepts more context: 1.05M tokens versus 500K.
Side by side
| GPT-6 Luna | Grok 4.6 | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 53.3 | 56.9 |
| Released | 2026-09-22 | 2026-08-12 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 500K |
| Max output | 128K | 500K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.50 | $6 |
| Results tracked | 42 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.6 leads
GPT-6 Luna: 55.5 (#25), Grok 4.6: 58.5 (#16)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| DeepSWE | 66.6% | 67.5% |
| FrontierCode | 42.4% | 48% |
| LMArena WebDev | 1581 | 1617 |
| SciCode | 54.6% | 56.5% |
| LMArena Coding | 1439 | 1465 |
| ALE-Bench | 1,577 | 1,508 |
| CursorBench | — | 41.4% |
| FrontierSWE | — | 25.3% |
| WeirdML | — | 67.3% |
Agentic & Tool Use Grok 4.6 leads
GPT-6 Luna: 33.3 (#54), Grok 4.6: 39.4 (#27)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| APEX-Agents | 44.3% | 65.3% |
| GDP.pdf | 23% | 17.2% |
| Vending-Bench 2 | — | 9,047 |
Reasoning Grok 4.6 leads
GPT-6 Luna: 48.2 (#41), Grok 4.6: 61.4 (#20)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 59.3% | 67.1% |
| NYT Connections (extended) | 68.7% | 80% |
| ARC-AGI-1 | 86.7% | 87.5% |
| CritPt | 19.4% | 19.7% |
| Chess Puzzles | 31% | 40% |
| LMArena Hard Prompts | 1411 | 1447 |
| Mystery Game Puzzles | 7% | 34% |
| DTBench | 90.1% | 97.3% |
| LMCA | 44.5% | 48.5% |
| Epoch Capabilities Index | 156.28 | 156.44 |
| SimpleBench | — | 75.9% |
| EBR-Bench | — | 30.5% |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Grok 4.6: 67.0 (#24)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 78.9% | 66% |
| FrontierMath Tier 4 | 56.1% | 31.7% |
| OTIS Mock AIME 2024-2025 | 98.9% | 99.2% |
| ProofBench | 64% | 51% |
| LMArena Math | 1416 | 1423 |
Knowledge Grok 4.6 leads
GPT-6 Luna: 57.0 (#41), Grok 4.6: 63.3 (#20)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 90.5% | 94% |
| SimpleQA Verified | 41.4% | 49.3% |
| LMArena Expert | 1444 | 1467 |
Multimodal Grok 4.6 leads
GPT-6 Luna: 42.4 (#30), Grok 4.6: 43.6 (#23)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1217 | 1263 |
| Blueprint-Bench 2 | 31.2% | 33.2% |
| Furniture Assembly | 44.2% | 40% |
| LMArena Document | — | 1452 |
Multilingual Grok 4.6 leads
GPT-6 Luna: 50.5 (#117), Grok 4.6: 53.0 (#74)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1386 | 1420 |
| LMArena Chinese | 1433 | 1480 |
| LMArena French | 1420 | 1461 |
| LMArena German | 1369 | 1431 |
| LMArena Japanese | 1369 | 1376 |
| LMArena Korean | 1360 | 1397 |
| LMArena Russian | 1394 | 1422 |
| LMArena Spanish | 1393 | 1404 |
Instruction Following Grok 4.6 leads
GPT-6 Luna: 74.3 (#99), Grok 4.6: 75.4 (#63)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1409 | 1431 |
Long Context Grok 4.6 leads
GPT-6 Luna: 43.0 (#111), Grok 4.6: 44.5 (#66)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1409 | 1454 |
Writing & Preference Grok 4.6 leads
GPT-6 Luna: 58.3 (#119), Grok 4.6: 62.3 (#80)
| Benchmark | GPT-6 Luna | Grok 4.6 |
|---|---|---|
| LMArena Text | 1391 | 1428 |
| LMArena Creative Writing | 1363 | 1428 |
| LMArena Multi-Turn | 1396 | 1425 |
Frequently asked questions
Is GPT-6 Luna better than Grok 4.6?
Grok 4.6 is the stronger model overall, scoring 56.9 to 53.3 on the Noometry Index. GPT-6 Luna costs 15× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.
Which is cheaper, GPT-6 Luna or Grok 4.6?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Grok 4.6 lists at $2 and $6.
Is GPT-6 Luna or Grok 4.6 better for coding?
Grok 4.6 scores higher on coding benchmarks: 58.5 versus 55.5 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 500K.
How many benchmarks do GPT-6 Luna and Grok 4.6 share?
42 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Grok 4.6 has 49.