Model comparison
Gemini 3.5 Flash Lite vs GPT-5.6 Luna
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 41.5 on the Noometry Index.
Last verified . 39 shared benchmarks.
GoogleSummary
- They share 39 benchmarks with published results for both. Gemini 3.5 Flash Lite scores higher in 1 category and GPT-5.6 Luna in 9 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 25.9.
- The biggest single-benchmark swing is FrontierMath Tier 4: 0% for Gemini 3.5 Flash Lite and 61% for GPT-5.6 Luna.
- GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $0.30 / $2.50 for Gemini 3.5 Flash Lite.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 1.05M.
Side by side
| Gemini 3.5 Flash Lite | GPT-5.6 Luna | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 41.5 | 54.6 |
| Released | 2026-07-21 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 128K |
| Input $ / M tokens | $0.30 | $0.20 |
| Output $ / M tokens | $2.50 | $1.20 |
| Results tracked | 39 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 42.4 (#104), GPT-5.6 Luna: 54.5 (#28)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena WebDev | 1440 | 1519 |
| SciCode | 41.3% | 53.6% |
| WeirdML | 39% | 60.9% |
| LMArena Coding | 1453 | 1466 |
| ALE-Bench | 765.27 | 1,667 |
| DeepSWE | — | 67.2% |
| FrontierCode | — | 39.8% |
| CursorBench | — | 35.9% |
Agentic & Tool Use GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 25.7 (#107), GPT-5.6 Luna: 34.4 (#45)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| APEX-Agents | 29.3% | 43% |
| GDP.pdf | 10% | 22.7% |
| BALROG | — | 45.6% |
| Vending-Bench 2 | — | 4,095 |
Reasoning GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 27.8 (#114), GPT-5.6 Luna: 47.6 (#43)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| ARC-AGI-2 | 10.3% | 59.5% |
| NYT Connections (extended) | 60.4% | 69.4% |
| ARC-AGI-1 | 53.5% | 88% |
| CritPt | 0% | 20.6% |
| Chess Puzzles | 22% | 40% |
| LMArena Hard Prompts | 1441 | 1451 |
| Mystery Game Puzzles | 19% | 21% |
| DTBench | 83.5% | 89.1% |
| LMCA | 37.6% | 48.5% |
| Epoch Capabilities Index | 145.13 | 156.39 |
| SimpleBench | — | 46.8% |
| Kagi LLM Benchmark | — | 49.1% |
| Surface Evolver Bench | — | 61.9% |
Math GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 25.9 (#264), GPT-5.6 Luna: 77.7 (#14)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 26% | 82.1% |
| FrontierMath Tier 4 | 0% | 61% |
| OTIS Mock AIME 2024-2025 | 71.1% | 98.3% |
| ProofBench | 13% | 60% |
| LMArena Math | 1433 | 1458 |
Knowledge GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 50.1 (#72), GPT-5.6 Luna: 58.5 (#34)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| GPQA Diamond | 83.3% | 91.6% |
| LMArena Expert | 1435 | 1478 |
| SimpleQA Verified | — | 41% |
Multimodal GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 41.1 (#40), GPT-5.6 Luna: 42.7 (#28)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena Vision | 1268 | 1258 |
| Blueprint-Bench 2 | — | 22.6% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1457 |
Multilingual Too close to call
Gemini 3.5 Flash Lite: 53.5 (#63), GPT-5.6 Luna: 52.8 (#78)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena Non-English | 1427 | 1417 |
| LMArena Chinese | 1469 | 1470 |
| LMArena French | 1451 | 1456 |
| LMArena German | 1454 | 1454 |
| LMArena Japanese | 1427 | 1411 |
| LMArena Korean | 1407 | 1415 |
| LMArena Russian | 1443 | 1428 |
| LMArena Spanish | 1442 | 1448 |
Instruction Following Too close to call
Gemini 3.5 Flash Lite: 74.8 (#85), GPT-5.6 Luna: 75.6 (#57)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena Instruction Following | 1420 | 1437 |
Long Context Too close to call
Gemini 3.5 Flash Lite: 43.9 (#84), GPT-5.6 Luna: 43.9 (#82)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena Longer Query | 1435 | 1436 |
Writing & Preference GPT-5.6 Luna leads
Gemini 3.5 Flash Lite: 64.2 (#57), GPT-5.6 Luna: 68.0 (#29)
| Benchmark | Gemini 3.5 Flash Lite | GPT-5.6 Luna |
|---|---|---|
| LMArena Text | 1435 | 1431 |
| LMArena Creative Writing | 1420 | 1396 |
| EQ-Bench Creative Writing | 1559 | 1829 |
| LMArena Multi-Turn | 1445 | 1434 |
| EQ-Bench 4 | — | 1156 |
Frequently asked questions
Is Gemini 3.5 Flash Lite better than GPT-5.6 Luna?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 41.5 on the Noometry Index.
Which is cheaper, Gemini 3.5 Flash Lite or GPT-5.6 Luna?
GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; Gemini 3.5 Flash Lite lists at $0.30 and $2.50.
Is Gemini 3.5 Flash Lite or GPT-5.6 Luna better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 1.05M.
How many benchmarks do Gemini 3.5 Flash Lite and GPT-5.6 Luna share?
39 benchmarks have published results for both models. Gemini 3.5 Flash Lite has 39 scored results on Noometry and GPT-5.6 Luna has 52.