Model comparison
Gemini 2.5 Flash vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 1.7× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Google
StepFunSummary
- They share 15 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 2 categories and Step 5 Preview in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 18.1.
- The biggest single-benchmark swing is CritPt: 1.1% for Gemini 2.5 Flash and 20.9% for Step 5 Preview.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Gemini 2.5 Flash accepts more context: 1.05M tokens versus 1.02M.
Side by side
| Gemini 2.5 Flash | Step 5 Preview | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 39.3 | 47.9 |
| Released | 2025-04-17 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.02M |
| Max output | 66K | 66K |
| Input $ / M tokens | $0.30 | $1 |
| Output $ / M tokens | $2.50 | $2.70 |
| Results tracked | 54 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Gemini 2.5 Flash: 35.8 (#220), Step 5 Preview: 51.8 (#36)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1424 | 1480 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 41.9% | — |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Not comparable
Gemini 2.5 Flash: 30.8 (#74), Step 5 Preview: —
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |
Reasoning Step 5 Preview leads
Gemini 2.5 Flash: 18.1 (#286), Step 5 Preview: 40.0 (#57)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| CritPt | 1.1% | 20.9% |
| LMArena Hard Prompts | 1422 | 1465 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| ARC-AGI-1 | 33.3% | — |
| EnigmaEval | 2.7% | — |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |
| Epoch Capabilities Index | 143.03 | — |
| ForecastBench | 60.6 | — |
Math Step 5 Preview leads
Gemini 2.5 Flash: 39.9 (#98), Step 5 Preview: 46.1 (#72)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Math | 1415 | 1470 |
| OTIS Mock AIME 2024-2025 | 73.1% | — |
| ProofBench | — | 42% |
| Omni-MATH | 38.5% | — |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Step 5 Preview leads
Gemini 2.5 Flash: 36.4 (#168), Step 5 Preview: 41.2 (#112)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1426 | 1470 |
| Humanity's Last Exam | 12.1% | — |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
Multimodal Too close to call
Gemini 2.5 Flash: 41.8 (#32), Step 5 Preview: 41.0 (#41)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1253 | 1267 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Step 5 Preview leads
Gemini 2.5 Flash: 52.3 (#88), Step 5 Preview: 53.8 (#54)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1409 | 1432 |
| LMArena Chinese | 1450 | 1519 |
| LMArena Russian | 1415 | 1436 |
| LMArena Spanish | 1421 | 1439 |
| LMArena French | 1433 | — |
| LMArena German | 1418 | — |
| LMArena Japanese | 1405 | — |
| LMArena Korean | 1385 | — |
Instruction Following Too close to call
Gemini 2.5 Flash: 75.7 (#54), Step 5 Preview: 76.0 (#50)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1405 | 1444 |
| IFEval | 89.8% | — |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), Step 5 Preview: 44.5 (#64)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1419 | 1455 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference Step 5 Preview leads
Gemini 2.5 Flash: 53.8 (#157), Step 5 Preview: 62.8 (#71)
| Benchmark | Gemini 2.5 Flash | Step 5 Preview |
|---|---|---|
| LMArena Text | 1417 | 1442 |
| LMArena Creative Writing | 1400 | 1410 |
| LMArena Multi-Turn | 1408 | 1447 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 1.7× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, Gemini 2.5 Flash or Step 5 Preview?
Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Gemini 2.5 Flash or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 35.8 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash does, with 1.05M tokens against 1.02M.
How many benchmarks do Gemini 2.5 Flash and Step 5 Preview share?
15 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Step 5 Preview has 18.