Model comparison
Gemini 3.6 Flash vs Gemini 3 Flash Preview
Gemini 3.6 Flash is the stronger model overall, scoring 54.1 to 52.3 on the Noometry Index.
Last verified . 39 shared benchmarks.
GoogleSummary
- They share 39 benchmarks with published results for both. Gemini 3.6 Flash scores higher in 7 categories and Gemini 3 Flash Preview in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.6 Flash leads 58.8 to 49.2.
- The biggest single-benchmark swing is Blueprint-Bench 2: 31.2% for Gemini 3.6 Flash and 0% for Gemini 3 Flash Preview.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.6 Flash.
Side by side
| Gemini 3.6 Flash | Gemini 3 Flash Preview | |
|---|---|---|
| Provider | ||
| Noometry Index | 54.1 | 52.3 |
| Released | 2026-07-21 | 2025-12-17 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 66K | 66K |
| Input $ / M tokens | $0.75 | $0.50 |
| Output $ / M tokens | $3.75 | $3 |
| Results tracked | 46 | 59 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3.6 Flash: 50.0 (#48), Gemini 3 Flash Preview: 50.9 (#42)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena WebDev | 1538 | 1439 |
| WeirdML | 56.1% | 61.6% |
| LMArena Coding | 1491 | 1460 |
| ALE-Bench | 715.52 | 1,367 |
| SWE-bench Verified | — | 75.4% |
| DeepSWE | 46.7% | — |
| FrontierCode | 34.4% | — |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 72.7% |
| SciCode | 52.7% | — |
| GSO | — | 9.8% |
Agentic & Tool Use Gemini 3 Flash Preview leads
Gemini 3.6 Flash: 32.3 (#65), Gemini 3 Flash Preview: 38.7 (#29)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| GDP.pdf | 14% | 10% |
| Terminal-Bench | — | 64.3% |
| APEX-Agents | 46.9% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 27.3% |
| τ²-bench Retail | — | 76.8% |
| τ²-bench Telecom | — | 91.2% |
| DeepResearch Bench | — | 49.8% |
| BALROG | — | 48.1% |
| LMArena Search | — | 1198 |
| Vending-Bench 2 | — | 3,635 |
Reasoning Gemini 3.6 Flash leads
Gemini 3.6 Flash: 58.8 (#22), Gemini 3 Flash Preview: 49.2 (#37)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| ARC-AGI-2 | 60.4% | 33.6% |
| NYT Connections (extended) | 89% | 83.1% |
| ARC-AGI-1 | 91.2% | 84.7% |
| Chess Puzzles | 43% | 40% |
| LMArena Hard Prompts | 1485 | 1465 |
| Mystery Game Puzzles | 30% | 26% |
| DTBench | 95.5% | 89.1% |
| LMCA | 44.9% | 43.1% |
| Epoch Capabilities Index | 154.25 | 151.8 |
| SimpleBench | — | 61.1% |
| CritPt | 10.6% | — |
| ForecastBench | — | 58.5 |
Math Gemini 3.6 Flash leads
Gemini 3.6 Flash: 57.3 (#40), Gemini 3 Flash Preview: 51.7 (#55)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| FrontierMath (Tiers 1-3) | 58.9% | 51.2% |
| FrontierMath Tier 4 | 22% | 17.1% |
| MathArena Final-Answer Competitions | 70.8% | 67.6% |
| OTIS Mock AIME 2024-2025 | 94.2% | 95.6% |
| ProofBench | 36% | 15% |
| LMArena Math | 1505 | 1473 |
| FrontierMath (Feb 2025 set) | — | 35.6% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Gemini 3.6 Flash leads
Gemini 3.6 Flash: 67.8 (#8), Gemini 3 Flash Preview: 58.8 (#33)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| GPQA Diamond | 94.1% | 89.4% |
| SimpleQA Verified | 66.2% | 66.8% |
| LMArena Expert | 1488 | 1462 |
| Vectara Hallucination Rate | — | 13.5% |
Multimodal Gemini 3 Flash Preview leads
Gemini 3.6 Flash: 38.5 (#64), Gemini 3 Flash Preview: 45.5 (#16)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Vision | 1298 | 1285 |
| Blueprint-Bench 2 | 31.2% | 0% |
| LMArena Document | 1456 | 1413 |
| GeoBench | — | 88% |
| VPCT | — | 72.6% |
| Furniture Assembly | 23.3% | — |
Multilingual Too close to call
Gemini 3.6 Flash: 56.5 (#19), Gemini 3 Flash Preview: 55.7 (#27)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Non-English | 1469 | 1458 |
| LMArena Chinese | 1531 | 1511 |
| LMArena French | 1504 | 1477 |
| LMArena German | 1478 | 1497 |
| LMArena Japanese | 1476 | 1489 |
| LMArena Korean | 1431 | 1443 |
| LMArena Russian | 1487 | 1480 |
| LMArena Spanish | 1475 | 1469 |
Instruction Following Gemini 3.6 Flash leads
Gemini 3.6 Flash: 77.0 (#33), Gemini 3 Flash Preview: 75.7 (#56)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Instruction Following | 1466 | 1437 |
Long Context Too close to call
Gemini 3.6 Flash: 45.1 (#50), Gemini 3 Flash Preview: 44.4 (#67)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Longer Query | 1474 | 1452 |
Writing & Preference Gemini 3.6 Flash leads
Gemini 3.6 Flash: 68.2 (#27), Gemini 3 Flash Preview: 65.5 (#45)
| Benchmark | Gemini 3.6 Flash | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Text | 1479 | 1466 |
| LMArena Creative Writing | 1465 | 1457 |
| LMArena Multi-Turn | 1481 | 1471 |
| EQ-Bench Creative Writing | 1604 | — |
Frequently asked questions
Is Gemini 3.6 Flash better than Gemini 3 Flash Preview?
Gemini 3.6 Flash is the stronger model overall, scoring 54.1 to 52.3 on the Noometry Index.
Which is cheaper, Gemini 3.6 Flash or Gemini 3 Flash Preview?
Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Gemini 3.6 Flash lists at $0.75 and $3.75.
Is Gemini 3.6 Flash or Gemini 3 Flash Preview better for coding?
They score almost the same on coding (50.0 vs 50.9); test both on your own repository before choosing.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do Gemini 3.6 Flash and Gemini 3 Flash Preview share?
39 benchmarks have published results for both models. Gemini 3.6 Flash has 46 scored results on Noometry and Gemini 3 Flash Preview has 59.