Model comparison
Gemma 3 27B vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 43.8 to 30.5 on the Noometry Index. Gemma 3 27B costs 14× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Google
StepFunSummary
- They share 2 benchmarks with published results for both. Gemma 3 27B scores higher in 0 categories and Step 5 Preview in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 5 Preview leads 50.6 to 22.1.
- The biggest single-benchmark swing is SciCode: 21.2% for Gemma 3 27B and 58.9% for Step 5 Preview.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 131K.
- Gemma 3 27B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 27B | Step 5 Preview | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 30.5 | 43.8 |
| Released | 2025-03-11 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.02M |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.08 | $1 |
| Output $ / M tokens | $0.16 | $2.70 |
| Results tracked | 43 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Gemma 3 27B: 22.1 (#329), Step 5 Preview: 50.6 (#40)
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| SciCode | 21.2% | 58.9% |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1569 |
| LiveBench Coding | 39.9% | — |
| LMArena Coding | 1322 | — |
Agentic & Tool Use Not comparable
Gemma 3 27B: 24.9 (#105), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
Reasoning Step 5 Preview leads
Gemma 3 27B: 16.3 (#297), Step 5 Preview: 38.0 (#58)
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| CritPt | 0% | 20.9% |
| Kagi LLM Benchmark | 40.4% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| LMArena Hard Prompts | 1340 | — |
| DTBench | 52.5% | — |
| LiveBench Data Analysis | 51.5% | — |
| LMCA | 12.3% | — |
| Epoch Capabilities Index | 130.04 | — |
| LiveBench | 50% | — |
Math Not comparable
Gemma 3 27B: 24.4 (#261), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | — |
| LiveBench Math | 55.4% | — |
| LMArena Math | 1312 | — |
| MATH Level 5 | 74% | — |
Knowledge Not comparable
Gemma 3 27B: 25.4 (#256), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| GPQA Diamond | 47.7% | — |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
| LMArena Expert | 1303 | — |
Multimodal Not comparable
Gemma 3 27B: 32.0 (#94), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1165 | — |
| GeoBench | 52% | — |
Multilingual Not comparable
Gemma 3 27B: 47.0 (#150), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1334 | — |
| LMArena Chinese | 1347 | — |
| LMArena French | 1368 | — |
| LMArena German | 1362 | — |
| LMArena Japanese | 1287 | — |
| LMArena Korean | 1308 | — |
| LMArena Russian | 1349 | — |
| LMArena Spanish | 1348 | — |
Instruction Following Not comparable
Gemma 3 27B: 70.8 (#155), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| LiveBench Instruction Following | 74.9% | — |
| LMArena Instruction Following | 1321 | — |
Long Context Not comparable
Gemma 3 27B: 27.6 (#288), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| Fiction.LiveBench | 33.3% | — |
| LMArena Longer Query | 1333 | — |
Writing & Preference Not comparable
Gemma 3 27B: 52.7 (#161), Step 5 Preview: —
| Benchmark | Gemma 3 27B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1358 | — |
| LMArena Creative Writing | 1346 | — |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench Creative Writing | 1266 | — |
| LMArena Multi-Turn | 1344 | — |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 43.8 to 30.5 on the Noometry Index. Gemma 3 27B costs 14× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 27B or Step 5 Preview?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Gemma 3 27B or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 50.6 versus 22.1 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 131K.
How many benchmarks do Gemma 3 27B and Step 5 Preview share?
2 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Step 5 Preview has 3.