Model comparison
Gemma 3n E4b IT vs GPT-5.4 nano
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 36.7 on the Noometry Index. Gemma 3n E4b IT costs 6.2× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
GoogleSummary
- They share 18 benchmarks with published results for both. Gemma 3n E4b IT scores higher in 0 categories and GPT-5.4 nano in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 nano leads 40.7 to 32.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 31.5% for Gemma 3n E4b IT and 39.7% for GPT-5.4 nano.
- Gemma 3n E4b IT is cheaper at $0.06 / $0.12 per million input/output tokens, against $0.20 / $1.25 for GPT-5.4 nano.
- GPT-5.4 nano accepts more context: 400K tokens versus 33K.
- Gemma 3n E4b IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 3n E4b IT | GPT-5.4 nano | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 36.7 | 41.9 |
| Released | 2025-05-20 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | 33K | 400K |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.06 | $0.20 |
| Output $ / M tokens | $0.12 | $1.25 |
| Results tracked | 18 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
Gemma 3n E4b IT: 36.4 (#193), GPT-5.4 nano: 43.1 (#82)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Coding | 1269 | 1404 |
| SciCode | — | 46.9% |
| WeirdML | — | 49.2% |
| ALE-Bench | — | 1,005 |
Reasoning GPT-5.4 nano leads
Gemma 3n E4b IT: 19.2 (#239), GPT-5.4 nano: 23.8 (#159)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| Kagi LLM Benchmark | 31.5% | 39.7% |
| LMArena Hard Prompts | 1284 | 1380 |
| ARC-AGI-2 | — | 5.7% |
| ARC-AGI-1 | — | 51.5% |
| CritPt | — | 9.3% |
| Chess Puzzles | — | 30% |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LMCA | — | 36.9% |
| Epoch Capabilities Index | — | 145.78 |
| ForecastBench | — | 57.3 |
Math GPT-5.4 nano leads
Gemma 3n E4b IT: 32.5 (#186), GPT-5.4 nano: 40.7 (#76)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Math | 1251 | 1406 |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| OTIS Mock AIME 2024-2025 | — | 87.8% |
| ProofBench | — | 5% |
| FrontierMath (Feb 2025 set) | — | 25.9% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5.4 nano leads
Gemma 3n E4b IT: 33.6 (#193), GPT-5.4 nano: 41.8 (#94)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Expert | 1245 | 1397 |
| GPQA Diamond | — | 78.5% |
| SimpleQA Verified | — | 11.7% |
| Vectara Hallucination Rate | — | 3.1% |
Multimodal Not comparable
Gemma 3n E4b IT: —, GPT-5.4 nano: 36.0 (#73)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | — | 1196 |
Multilingual GPT-5.4 nano leads
Gemma 3n E4b IT: 43.5 (#179), GPT-5.4 nano: 48.8 (#136)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1285 | 1359 |
| LMArena Chinese | 1310 | 1391 |
| LMArena French | 1330 | 1396 |
| LMArena German | 1312 | 1368 |
| LMArena Japanese | 1274 | 1342 |
| LMArena Korean | 1259 | 1321 |
| LMArena Russian | 1288 | 1363 |
| LMArena Spanish | 1305 | 1368 |
Instruction Following GPT-5.4 nano leads
Gemma 3n E4b IT: 66.3 (#205), GPT-5.4 nano: 72.1 (#139)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1255 | 1362 |
Long Context GPT-5.4 nano leads
Gemma 3n E4b IT: 38.8 (#185), GPT-5.4 nano: 41.6 (#131)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1277 | 1365 |
Writing & Preference GPT-5.4 nano leads
Gemma 3n E4b IT: 50.3 (#181), GPT-5.4 nano: 55.9 (#137)
| Benchmark | Gemma 3n E4b IT | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1306 | 1372 |
| LMArena Creative Writing | 1287 | 1313 |
| LMArena Multi-Turn | 1276 | 1382 |
Frequently asked questions
Is Gemma 3n E4b IT better than GPT-5.4 nano?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 36.7 on the Noometry Index. Gemma 3n E4b IT costs 6.2× less per token, which makes it the better buy when GPT-5.4 nano's lead doesn't matter for your workload.
Which is cheaper, Gemma 3n E4b IT or GPT-5.4 nano?
Gemma 3n E4b IT is cheaper. It lists at $0.06 per million input tokens and $0.12 per million output tokens; GPT-5.4 nano lists at $0.20 and $1.25.
Is Gemma 3n E4b IT or GPT-5.4 nano better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.1 versus 36.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 33K.
How many benchmarks do Gemma 3n E4b IT and GPT-5.4 nano share?
18 benchmarks have published results for both models. Gemma 3n E4b IT has 18 scored results on Noometry and GPT-5.4 nano has 40.