Model comparison
Gemma 3 4B vs GPT-4.1 mini
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 28.1 on the Noometry Index. Gemma 3 4B costs 14× less per token, which makes it the better buy when GPT-4.1 mini's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
GoogleSummary
- They share 21 benchmarks with published results for both. Gemma 3 4B scores higher in 3 categories and GPT-4.1 mini in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-4.1 mini leads 34.7 to 11.8.
- The biggest single-benchmark swing is GPQA Diamond: 23.2% for Gemma 3 4B and 65.8% for GPT-4.1 mini.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.40 / $1.60 for GPT-4.1 mini.
- GPT-4.1 mini 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 | GPT-4.1 mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 28.1 | 33.6 |
| Released | 2025-03-12 | 2025-04-14 |
| Weights | Open | Proprietary |
| Context window | 131K | 1.05M |
| Max output | 4K | 33K |
| Input $ / M tokens | $0.04 | $0.40 |
| Output $ / M tokens | $0.08 | $1.60 |
| Results tracked | 22 | 47 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
Gemma 3 4B: 35.9 (#215), GPT-4.1 mini: 30.6 (#293)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Coding | 1230 | 1367 |
| SWE-bench Verified (bash only) | — | 23.9% |
| Aider Polyglot | — | 32.4% |
| SciCode | — | 40.4% |
| WeirdML | — | 37.6% |
| BigCodeBench Instruct | — | 48.9% |
| CadEval | — | 16% |
Agentic & Tool Use GPT-4.1 mini leads
Gemma 3 4B: 20.9 (#142), GPT-4.1 mini: 33.3 (#55)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | 50.5% |
Reasoning Gemma 3 4B leads
Gemma 3 4B: 13.2 (#335), GPT-4.1 mini: 10.8 (#340)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| Kagi LLM Benchmark | 25.2% | 48.6% |
| Chess Puzzles | 0% | 7% |
| LMArena Hard Prompts | 1253 | 1349 |
| DTBench | 50.9% | 68.8% |
| LMCA | 2.8% | 21.1% |
| Epoch Capabilities Index | 116.02 | 135.01 |
| ARC-AGI-2 | — | 0% |
| ARC-AGI-1 | — | 3.5% |
| CritPt | — | 0% |
| Mystery Game Puzzles | — | 7% |
Math GPT-4.1 mini leads
Gemma 3 4B: 16.8 (#292), GPT-4.1 mini: 24.1 (#270)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 7.5% | 44.7% |
| LMArena Math | 1239 | 1343 |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| Omni-MATH | — | 49.1% |
| MATH Level 5 | — | 87.3% |
| FrontierMath (Feb 2025 set) | — | 4.5% |
Knowledge GPT-4.1 mini leads
Gemma 3 4B: 11.8 (#299), GPT-4.1 mini: 34.7 (#194)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 23.2% | 65.8% |
| LMArena Expert | 1223 | 1338 |
| SimpleQA Verified | — | 12.7% |
| MMLU-Pro | — | 78.3% |
| Vectara Hallucination Rate | 6.4% | — |
| GPQA (HELM) | — | 61.4% |
Multimodal Not comparable
Gemma 3 4B: —, GPT-4.1 mini: 35.8 (#82)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | — | 1181 |
Multilingual GPT-4.1 mini leads
Gemma 3 4B: 42.5 (#194), GPT-4.1 mini: 45.7 (#166)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1273 | 1318 |
| LMArena German | 1281 | 1351 |
| LMArena Russian | 1294 | 1324 |
| LMArena Chinese | — | 1329 |
| LMArena French | — | 1358 |
| LMArena Japanese | — | 1290 |
| LMArena Korean | — | 1298 |
| LMArena Spanish | — | 1319 |
Instruction Following GPT-4.1 mini leads
Gemma 3 4B: 65.2 (#225), GPT-4.1 mini: 73.7 (#118)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Instruction Following | 1239 | 1333 |
| IFEval | — | 90.4% |
Long Context Gemma 3 4B leads
Gemma 3 4B: 38.7 (#194), GPT-4.1 mini: 31.8 (#275)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Longer Query | 1273 | 1344 |
| Fiction.LiveBench | — | 44.4% |
Writing & Preference GPT-4.1 mini leads
Gemma 3 4B: 42.0 (#239), GPT-4.1 mini: 48.6 (#199)
| Benchmark | Gemma 3 4B | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1291 | 1340 |
| LMArena Creative Writing | 1271 | 1300 |
| EQ-Bench Creative Writing | 1068 | 1147 |
| LMArena Multi-Turn | 1255 | 1354 |
| WildBench | — | 83.8% |
Frequently asked questions
Is Gemma 3 4B better than GPT-4.1 mini?
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 28.1 on the Noometry Index. Gemma 3 4B costs 14× less per token, which makes it the better buy when GPT-4.1 mini's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or GPT-4.1 mini?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; GPT-4.1 mini lists at $0.40 and $1.60.
Is Gemma 3 4B or GPT-4.1 mini better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 30.6 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 mini does, with 1.05M tokens against 131K.
How many benchmarks do Gemma 3 4B and GPT-4.1 mini share?
21 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and GPT-4.1 mini has 47.