Model comparison
Gemini 3 Flash Preview vs Step 5 Preview
Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 47.9 on the Noometry Index.
Last verified . 16 shared benchmarks.
Google
StepFunSummary
- They share 16 benchmarks with published results for both. Gemini 3 Flash Preview scores higher in 6 categories and Step 5 Preview in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3 Flash Preview leads 58.8 to 41.2.
- The biggest single-benchmark swing is ProofBench: 15% for Gemini 3 Flash Preview and 42% for Step 5 Preview.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Gemini 3 Flash Preview accepts more context: 1.05M tokens versus 1.02M.
Side by side
| Gemini 3 Flash Preview | Step 5 Preview | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 52.3 | 47.9 |
| Released | 2025-12-17 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.02M |
| Max output | 66K | 66K |
| Input $ / M tokens | $0.50 | $1 |
| Output $ / M tokens | $3 | $2.70 |
| Results tracked | 59 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 3 Flash Preview: 50.9 (#42), Step 5 Preview: 51.8 (#36)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1439 | 1564 |
| LMArena Coding | 1460 | 1480 |
| SWE-bench Verified | 75.4% | — |
| SWE-bench Verified (bash only) | 75.8% | — |
| SWE-bench Multilingual | 72.7% | — |
| SciCode | — | 58.9% |
| GSO | 9.8% | — |
| WeirdML | 61.6% | — |
| ALE-Bench | 1,367 | — |
Agentic & Tool Use Not comparable
Gemini 3 Flash Preview: 38.7 (#29), Step 5 Preview: —
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 64.3% | — |
| 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% | — |
| GDP.pdf | 10% | — |
| LMArena Search | 1198 | — |
| Vending-Bench 2 | 3,635 | — |
Reasoning Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 49.2 (#37), Step 5 Preview: 40.0 (#57)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1465 | 1465 |
| ARC-AGI-2 | 33.6% | — |
| SimpleBench | 61.1% | — |
| NYT Connections (extended) | 83.1% | — |
| ARC-AGI-1 | 84.7% | — |
| CritPt | — | 20.9% |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 26% | — |
| DTBench | 89.1% | — |
| LMCA | 43.1% | — |
| Epoch Capabilities Index | 151.8 | — |
| ForecastBench | 58.5 | — |
Math Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 51.7 (#55), Step 5 Preview: 46.1 (#72)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| ProofBench | 15% | 42% |
| LMArena Math | 1473 | 1470 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 17.1% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| OTIS Mock AIME 2024-2025 | 95.6% | — |
| FrontierMath (Feb 2025 set) | 35.6% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 58.8 (#33), Step 5 Preview: 41.2 (#112)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1462 | 1470 |
| GPQA Diamond | 89.4% | — |
| SimpleQA Verified | 66.8% | — |
| Vectara Hallucination Rate | 13.5% | — |
Multimodal Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 45.5 (#16), Step 5 Preview: 41.0 (#41)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1285 | 1267 |
| GeoBench | 88% | — |
| VPCT | 72.6% | — |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1413 | — |
Multilingual Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 55.7 (#27), Step 5 Preview: 53.8 (#54)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1458 | 1432 |
| LMArena Chinese | 1511 | 1519 |
| LMArena Russian | 1480 | 1436 |
| LMArena Spanish | 1469 | 1439 |
| LMArena French | 1477 | — |
| LMArena German | 1497 | — |
| LMArena Japanese | 1489 | — |
| LMArena Korean | 1443 | — |
Instruction Following Too close to call
Gemini 3 Flash Preview: 75.7 (#56), Step 5 Preview: 76.0 (#50)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1437 | 1444 |
Long Context Too close to call
Gemini 3 Flash Preview: 44.4 (#67), Step 5 Preview: 44.5 (#64)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1452 | 1455 |
Writing & Preference Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 65.5 (#45), Step 5 Preview: 62.8 (#71)
| Benchmark | Gemini 3 Flash Preview | Step 5 Preview |
|---|---|---|
| LMArena Text | 1466 | 1442 |
| LMArena Creative Writing | 1457 | 1410 |
| LMArena Multi-Turn | 1471 | 1447 |
Frequently asked questions
Is Gemini 3 Flash Preview better than Step 5 Preview?
Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 47.9 on the Noometry Index.
Which is cheaper, Gemini 3 Flash Preview or Step 5 Preview?
Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Gemini 3 Flash Preview or Step 5 Preview better for coding?
They score almost the same on coding (50.9 vs 51.8); test both on your own repository before choosing.
Which has the bigger context window?
Gemini 3 Flash Preview does, with 1.05M tokens against 1.02M.
How many benchmarks do Gemini 3 Flash Preview and Step 5 Preview share?
16 benchmarks have published results for both models. Gemini 3 Flash Preview has 59 scored results on Noometry and Step 5 Preview has 18.