Model comparison
Claude Opus 5 vs GPT-6 Luna
Claude Opus 5 is the stronger model overall, scoring 67.7 to 45.1 on the Noometry Index. GPT-6 Luna costs 50× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. Claude Opus 5 scores higher in 9 categories and GPT-6 Luna in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5 leads 86.1 to 35.6.
- The biggest single-benchmark swing is NYT Connections (extended): 94.3% for Claude Opus 5 and 68.7% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- GPT-6 Luna accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Opus 5 | GPT-6 Luna | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 67.7 | 45.1 |
| Released | 2026-07-24 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $5 | $0.10 |
| Output $ / M tokens | $25 | $0.50 |
| Results tracked | 57 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 5 leads
Claude Opus 5: 67.0 (#5), GPT-6 Luna: 54.7 (#24)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| DeepSWE | 73.6% | 66.6% |
| FrontierCode | 53.4% | 42.4% |
| LMArena WebDev | 1694 | 1583 |
| LMArena Coding | 1533 | 1448 |
| CursorBench | 46.6% | — |
| FrontierSWE | 52% | — |
| SciCode | 56.4% | — |
| WeirdML | 91.8% | — |
| ALE-Bench | 2,165 | — |
Agentic & Tool Use Not comparable
Claude Opus 5: 55.5 (#1), GPT-6 Luna: —
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| Vending-Bench 2 | 11,182 | — |
Reasoning Claude Opus 5 leads
Claude Opus 5: 77.1 (#3), GPT-6 Luna: 31.7 (#68)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| NYT Connections (extended) | 94.3% | 68.7% |
| LMArena Hard Prompts | 1525 | 1415 |
| ARC-AGI-2 | 90.4% | — |
| SimpleBench | 80.6% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 29.1% | — |
| Chess Puzzles | 42% | — |
| EBR-Bench | 45.7% | — |
| Mystery Game Puzzles | 59% | — |
| DTBench | 97.9% | — |
| LMCA | 64.5% | — |
| Bench to the Future 3 | 0.12 | — |
| Epoch Capabilities Index | 162.67 | — |
Math Claude Opus 5 leads
Claude Opus 5: 86.1 (#6), GPT-6 Luna: 35.6 (#140)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Math | 1533 | 1396 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 99% | — |
Knowledge Claude Opus 5 leads
Claude Opus 5: 66.8 (#8), GPT-6 Luna: 40.1 (#110)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Expert | 1558 | 1458 |
| GPQA Diamond | 93.9% | — |
| SimpleQA Verified | 59.9% | — |
Multimodal Claude Opus 5 leads
Claude Opus 5: 51.1 (#6), GPT-6 Luna: 42.0 (#27)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| Furniture Assembly | 60.8% | 44.2% |
| LMArena Vision | 1321 | — |
| Blueprint-Bench 2 | 30.4% | — |
| LMArena Document | 1516 | — |
Multilingual Claude Opus 5 leads
Claude Opus 5: 58.8 (#3), GPT-6 Luna: 51.0 (#111)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1500 | 1390 |
| LMArena Chinese | 1575 | 1425 |
| LMArena Russian | 1507 | 1392 |
| LMArena Spanish | 1519 | 1416 |
| LMArena French | 1522 | — |
| LMArena German | 1522 | — |
| LMArena Japanese | 1510 | — |
| LMArena Korean | 1526 | — |
Instruction Following Claude Opus 5 leads
Claude Opus 5: 79.3 (#5), GPT-6 Luna: 74.7 (#85)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1516 | 1414 |
Long Context Claude Opus 5 leads
Claude Opus 5: 46.5 (#20), GPT-6 Luna: 43.3 (#95)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1516 | 1417 |
Writing & Preference Claude Opus 5 leads
Claude Opus 5: 79.3 (#2), GPT-6 Luna: 58.4 (#115)
| Benchmark | Claude Opus 5 | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1506 | 1392 |
| LMArena Creative Writing | 1491 | 1357 |
| LMArena Multi-Turn | 1496 | 1400 |
| EQ-Bench Creative Writing | 2133 | — |
| EQ-Bench 4 | 1385 | — |
Frequently asked questions
Is Claude Opus 5 better than GPT-6 Luna?
Claude Opus 5 is the stronger model overall, scoring 67.7 to 45.1 on the Noometry Index. GPT-6 Luna costs 50× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 5 or GPT-6 Luna?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Claude Opus 5 lists at $5 and $25.
Is Claude Opus 5 or GPT-6 Luna better for coding?
Claude Opus 5 scores higher on coding benchmarks: 67.0 versus 54.7 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 1M.
How many benchmarks do Claude Opus 5 and GPT-6 Luna share?
18 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GPT-6 Luna has 18.