Gemini 3.6 Flash vs GPT-5.6 Luna
GPT-5.6 Luna leads Gemini 3.6 Flash by 3.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Gemini 3.6 Flash wins 8 and GPT-5.6 Luna wins 10 of 18 shared benchmarks. GPT-5.6 Luna is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Gemini 3.6 Flash | GPT-5.6 Luna |
|---|---|---|
| Agentic coding | 53 | 59 |
| Long-horizon autonomy | 58 | 61 |
| Knowledge-work agents | 39 | 39 |
| Computer use | 37 | 40 |
| Security & context learning | 49 | 51 |
Shared benchmarks
| Benchmark | Gemini 3.6 Flash | GPT-5.6 Luna |
|---|---|---|
| APEX-Agentsindex | 46.9% | 43.0% |
| DeepSWEindex | 46.7% | 67.2% |
| ARC-AGI | 91.2% | 88.0% |
| ARC-AGI-2 | 60.4% | 59.5% |
| Chess Puzzles | 42.0% | 43.0% |
| CritPt | 10.6% | 20.6% |
| DTBench | 95.5% | 89.1% |
| FrontierCode | 34.4% | 39.8% |
| FrontierMath Tier 4 | 22.0% | 61.0% |
| FrontierMath Tiers 1-3 | 58.9% | 82.1% |
| GPQA Diamond | 94.1% | 91.6% |
| LMCA | 44.9% | 48.5% |
| MirrorCode | 23.3% | 21.1% |
| OTIS Mock AIME | 94.2% | 98.3% |
| ProofBench | 36.0% | 60.0% |
| SciCode | 53.4% | 53.6% |
| SimpleQA Verified | 66.2% | 41.0% |
| WebDev Arena | 1537 | 1519 |
Specs and price
| Gemini 3.6 Flash | GPT-5.6 Luna | |
|---|---|---|
| Agent Index | 48.3 (#39) | 51.4 (#34) |
| Agent $/M | $0.35 | $0.10 |
| Input / output $/M | $0.75 / $3.75 | $0.20 / $1.20 |
| Context | 1M | 1.1M |
| Released | Jul 21, 2026 | Jul 9, 2026 |
| Weights | Closed | Closed |
| API hosts | 6 | 5 |
Don’t pick one. Route.
Shogo can run Gemini 3.6 Flash 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