Claude Opus 5.5 vs GPT-5.6 Luna
Claude Opus 5.5 leads GPT-5.6 Luna by 19.5 points on the Shogo Agent Index. Claude Opus 5.5 wins 21 and GPT-5.6 Luna wins 0 of 21 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Opus 5.5 | GPT-5.6 Luna |
|---|---|---|
| Agentic coding | 81 | 59 |
| Long-horizon autonomy | 76 | 61 |
| Knowledge-work agents | 58 | 39 |
| Computer use | 66 | 40 |
| Security & context learning | 62 | 51 |
Shared benchmarks
| Benchmark | Claude Opus 5.5 | GPT-5.6 Luna |
|---|---|---|
| APEX-Agentsindex | 73.5% | 43.0% |
| Vending-Bench 2index | $9,235 | $4,095 |
| CursorBenchindex | 57.8% | 35.9% |
| ARC-AGI | 98.5% | 88.0% |
| ARC-AGI-2 | 93.3% | 59.5% |
| Chess Puzzles | 67.0% | 43.0% |
| CritPt | 31.7% | 20.6% |
| DTBench | 98.9% | 89.1% |
| FrontierCode | 54.6% | 39.8% |
| FrontierMath Tier 4 | 95.0% | 61.0% |
| FrontierMath Tiers 1-3 | 91.2% | 82.1% |
| GPQA Diamond | 94.1% | 91.6% |
| LMCA | 68.2% | 48.5% |
| MirrorCode | 95.6% | 21.1% |
| OTIS Mock AIME | 100.0% | 98.3% |
| ProofBench | 100.0% | 60.0% |
| SciCode | 66.9% | 53.6% |
| SimpleBench | 88.4% | 46.8% |
| SimpleQA Verified | 72.2% | 41.0% |
| WebDev Arena | 1820 | 1519 |
| WeirdML v3 | 31.2% | 5.2% |
Specs and price
| Claude Opus 5.5 | GPT-5.6 Luna | |
|---|---|---|
| Agent Index | 70.9 (#3) | 51.4 (#34) |
| Agent $/M | $1.69 | $0.10 |
| Input / output $/M | $4.00 / $20.00 | $0.20 / $1.20 |
| Context | 1M | 1.1M |
| Released | Sep 22, 2026 | Jul 9, 2026 |
| Weights | Closed | Closed |
| API hosts | 6 | 5 |
Don’t pick one. Route.
Shogo can run Claude Opus 5.5 and GPT-5.6 Luna side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo