Model comparison
Gemma 3 4B vs MiMo-V2-Pro
MiMo-V2-Pro is the stronger model overall, scoring 43.0 to 28.1 on the Noometry Index. Gemma 3 4B costs 11× less per token, which makes it the better buy when MiMo-V2-Pro's lead doesn't matter for your workload.
Last verified . 12 shared benchmarks.
GoogleSummary
- They share 12 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and MiMo-V2-Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiMo-V2-Pro leads 41.4 to 11.8.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.43 / $0.87 for MiMo-V2-Pro.
- MiMo-V2-Pro accepts more context: 1.05M tokens versus 131K.
- Gemma 3 4B has downloadable open weights; the other is API-only.
Side by side
| Gemma 3 4B | MiMo-V2-Pro | |
|---|---|---|
| Provider | Xiaomi | |
| Noometry Index | 28.1 | 43.0 |
| Released | 2025-03-12 | 2026-03-18 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.05M |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.04 | $0.43 |
| Output $ / M tokens | $0.08 | $0.87 |
| Results tracked | 22 | 23 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiMo-V2-Pro leads
Gemma 3 4B: 35.9 (#215), MiMo-V2-Pro: 43.8 (#83)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Coding | 1230 | 1476 |
| LMArena WebDev | — | 1433 |
| ALE-Bench | — | 785.17 |
Agentic & Tool Use Not comparable
Gemma 3 4B: 20.9 (#142), MiMo-V2-Pro: —
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
Reasoning MiMo-V2-Pro leads
Gemma 3 4B: 13.2 (#335), MiMo-V2-Pro: 22.1 (#206)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1457 |
| Kagi LLM Benchmark | 25.2% | — |
| NYT Connections (extended) | — | 25.8% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 45.9% |
| DTBench | 50.9% | — |
| LMCA | 2.8% | — |
| Epoch Capabilities Index | 116.02 | — |
Math MiMo-V2-Pro leads
Gemma 3 4B: 16.8 (#292), MiMo-V2-Pro: 39.5 (#102)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Math | 1239 | 1447 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
Knowledge MiMo-V2-Pro leads
Gemma 3 4B: 11.8 (#299), MiMo-V2-Pro: 41.4 (#111)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Expert | 1223 | 1478 |
| GPQA Diamond | 23.2% | — |
| Vectara Hallucination Rate | 6.4% | — |
Multilingual MiMo-V2-Pro leads
Gemma 3 4B: 42.5 (#194), MiMo-V2-Pro: 52.7 (#81)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Non-English | 1273 | 1416 |
| LMArena German | 1281 | 1417 |
| LMArena Russian | 1294 | 1427 |
| LMArena Chinese | — | 1456 |
| LMArena French | — | 1469 |
| LMArena Japanese | — | 1366 |
| LMArena Korean | — | 1400 |
| LMArena Spanish | — | 1457 |
Instruction Following MiMo-V2-Pro leads
Gemma 3 4B: 65.2 (#225), MiMo-V2-Pro: 76.0 (#49)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Instruction Following | 1239 | 1445 |
Long Context MiMo-V2-Pro leads
Gemma 3 4B: 38.7 (#194), MiMo-V2-Pro: 41.5 (#138)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Longer Query | 1273 | 1455 |
| CL-bench | — | 15.7% |
| CL-bench Life | — | 6.9% |
Writing & Preference MiMo-V2-Pro leads
Gemma 3 4B: 42.0 (#239), MiMo-V2-Pro: 62.8 (#70)
| Benchmark | Gemma 3 4B | MiMo-V2-Pro |
|---|---|---|
| LMArena Text | 1291 | 1436 |
| LMArena Creative Writing | 1271 | 1415 |
| LMArena Multi-Turn | 1255 | 1456 |
| EQ-Bench Creative Writing | 1068 | — |
Frequently asked questions
Is Gemma 3 4B better than MiMo-V2-Pro?
MiMo-V2-Pro is the stronger model overall, scoring 43.0 to 28.1 on the Noometry Index. Gemma 3 4B costs 11× less per token, which makes it the better buy when MiMo-V2-Pro's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or MiMo-V2-Pro?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; MiMo-V2-Pro lists at $0.43 and $0.87.
Is Gemma 3 4B or MiMo-V2-Pro better for coding?
MiMo-V2-Pro scores higher on coding benchmarks: 43.8 versus 35.9 in the Noometry coding category.
Which has the bigger context window?
MiMo-V2-Pro does, with 1.05M tokens against 131K.
How many benchmarks do Gemma 3 4B and MiMo-V2-Pro share?
12 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and MiMo-V2-Pro has 23.