Model comparison
Gemma 4 31B IT vs Step 3.5 Flash
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 42.3 on the Noometry Index.
Last verified . 15 shared benchmarks.
Google
StepFunSummary
- They share 15 benchmarks with published results for both. Gemma 4 31B IT scores higher in 6 categories and Step 3.5 Flash in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 4 31B IT leads 27.2 to 22.2.
- The biggest single-benchmark swing is NYT Connections (extended): 70.6% for Gemma 4 31B IT and 28.4% for Step 3.5 Flash.
- Both cost about the same: $0.09 input and $0.34 output per million tokens.
- Gemma 4 31B IT accepts more context: 262K tokens versus 256K.
Side by side
| Gemma 4 31B IT | Step 3.5 Flash | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 43.5 | 42.3 |
| Released | 2026-04-02 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 262K | 256K |
| Max output | 33K | 256K |
| Input $ / M tokens | $0.09 | $0.10 |
| Output $ / M tokens | $0.34 | $0.30 |
| Results tracked | 35 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 4 31B IT: 42.3 (#108), Step 3.5 Flash: 42.4 (#105)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1459 | 1436 |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
| ALE-Bench | 925.5 | — |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| NYT Connections (extended) | 70.6% | 28.4% |
| LMArena Hard Prompts | 1448 | 1411 |
| Kagi LLM Benchmark | 63.5% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
| Epoch Capabilities Index | 142.74 | — |
Math Too close to call
Gemma 4 31B IT: 43.2 (#81), Step 3.5 Flash: 42.6 (#84)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1465 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
Knowledge Step 3.5 Flash leads
Gemma 4 31B IT: 37.9 (#151), Step 3.5 Flash: 39.6 (#132)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1465 | 1421 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Step 3.5 Flash: —
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Step 3.5 Flash: 50.5 (#119)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1431 | 1385 |
| LMArena Chinese | 1476 | 1447 |
| LMArena French | 1435 | 1421 |
| LMArena Russian | 1460 | 1385 |
| LMArena Spanish | 1444 | 1419 |
| LMArena German | — | 1405 |
| LMArena Japanese | — | 1354 |
| LMArena Korean | — | 1352 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Step 3.5 Flash: 73.1 (#124)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1433 | 1385 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Step 3.5 Flash: 42.8 (#117)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1446 | 1402 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Step 3.5 Flash: 58.8 (#113)
| Benchmark | Gemma 4 31B IT | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1443 | 1403 |
| LMArena Creative Writing | 1415 | 1357 |
| LMArena Multi-Turn | 1452 | 1405 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Step 3.5 Flash?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 42.3 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; Gemma 4 31B IT lists at $0.09 and $0.34.
Is Gemma 4 31B IT or Step 3.5 Flash better for coding?
They score almost the same on coding (42.3 vs 42.4); test both on your own repository before choosing.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 256K.
How many benchmarks do Gemma 4 31B IT and Step 3.5 Flash share?
15 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Step 3.5 Flash has 19.