Model comparison
Gemma 3 4B vs Step 2 16k Exp 202412
Step 2 16k Exp 202412 is the stronger model overall, scoring 39.2 to 28.1 on the Noometry Index.
Last verified . 11 shared benchmarks.
Google
StepFunSummary
- They share 11 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and Step 2 16k Exp 202412 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 2 16k Exp 202412 leads 35.2 to 11.8.
- Gemma 3 4B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 4B | Step 2 16k Exp 202412 | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 28.1 | 39.2 |
| Released | 2025-03-12 | — |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.04 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 22 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 2 16k Exp 202412 leads
Gemma 3 4B: 35.9 (#215), Step 2 16k Exp 202412: 38.5 (#174)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Coding | 1230 | 1316 |
Agentic & Tool Use Not comparable
Gemma 3 4B: 20.9 (#142), Step 2 16k Exp 202412: —
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
Reasoning Step 2 16k Exp 202412 leads
Gemma 3 4B: 13.2 (#335), Step 2 16k Exp 202412: 25.9 (#142)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1299 |
| Kagi LLM Benchmark | 25.2% | — |
| Chess Puzzles | 0% | — |
| DTBench | 50.9% | — |
| LMCA | 2.8% | — |
| Epoch Capabilities Index | 116.02 | — |
Math Step 2 16k Exp 202412 leads
Gemma 3 4B: 16.8 (#292), Step 2 16k Exp 202412: 36.3 (#169)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Math | 1239 | 1304 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
Knowledge Step 2 16k Exp 202412 leads
Gemma 3 4B: 11.8 (#299), Step 2 16k Exp 202412: 35.2 (#187)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Expert | 1223 | 1279 |
| GPQA Diamond | 23.2% | — |
| Vectara Hallucination Rate | 6.4% | — |
Multilingual Step 2 16k Exp 202412 leads
Gemma 3 4B: 42.5 (#194), Step 2 16k Exp 202412: 43.7 (#181)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Non-English | 1273 | 1290 |
| LMArena Russian | 1294 | 1324 |
| LMArena Chinese | — | 1331 |
| LMArena German | 1281 | — |
Instruction Following Step 2 16k Exp 202412 leads
Gemma 3 4B: 65.2 (#225), Step 2 16k Exp 202412: 67.9 (#192)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Instruction Following | 1239 | 1287 |
Long Context Step 2 16k Exp 202412 leads
Gemma 3 4B: 38.7 (#194), Step 2 16k Exp 202412: 39.7 (#170)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Longer Query | 1273 | 1307 |
Writing & Preference Step 2 16k Exp 202412 leads
Gemma 3 4B: 42.0 (#239), Step 2 16k Exp 202412: 52.3 (#173)
| Benchmark | Gemma 3 4B | Step 2 16k Exp 202412 |
|---|---|---|
| LMArena Text | 1291 | 1321 |
| LMArena Creative Writing | 1271 | 1328 |
| LMArena Multi-Turn | 1255 | 1292 |
| EQ-Bench Creative Writing | 1068 | — |
Frequently asked questions
Is Gemma 3 4B better than Step 2 16k Exp 202412?
Step 2 16k Exp 202412 is the stronger model overall, scoring 39.2 to 28.1 on the Noometry Index.
Is Gemma 3 4B or Step 2 16k Exp 202412 better for coding?
Step 2 16k Exp 202412 scores higher on coding benchmarks: 38.5 versus 35.9 in the Noometry coding category.
How many benchmarks do Gemma 3 4B and Step 2 16k Exp 202412 share?
11 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Step 2 16k Exp 202412 has 12.