Model comparison
Gemma 4 26B A4B IT vs GPT-5.4 mini
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 43.1 on the Noometry Index. Gemma 4 26B A4B IT costs 14× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
GoogleSummary
- They share 28 benchmarks with published results for both. Gemma 4 26B A4B IT scores higher in 5 categories and GPT-5.4 mini in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 mini leads 30.3 to 21.1.
- The biggest single-benchmark swing is WeirdML: 35.2% for Gemma 4 26B A4B IT and 60.3% for GPT-5.4 mini.
- Gemma 4 26B A4B IT is cheaper at $0.0765 / $0.26 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 262K.
- Gemma 4 26B A4B IT has downloadable open weights; the other is API-only.
Side by side
| Gemma 4 26B A4B IT | GPT-5.4 mini | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 43.1 | 44.8 |
| Released | 2026-04-02 | 2026-03-17 |
| Weights | Open | Proprietary |
| Context window | 262K | 400K |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.0765 | $0.75 |
| Output $ / M tokens | $0.26 | $4.50 |
| Results tracked | 28 | 46 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
Gemma 4 26B A4B IT: 38.8 (#156), GPT-5.4 mini: 45.0 (#67)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena WebDev | 1359 | 1397 |
| SciCode | 40% | 49.9% |
| WeirdML | 35.2% | 60.3% |
| LMArena Coding | 1446 | 1438 |
| ALE-Bench | 927.17 | 1,189 |
| FrontierCode | — | 27% |
Agentic & Tool Use Not comparable
Gemma 4 26B A4B IT: —, GPT-5.4 mini: 29.6 (#74)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| DeepResearch Bench | — | 36.3% |
Reasoning GPT-5.4 mini leads
Gemma 4 26B A4B IT: 21.1 (#205), GPT-5.4 mini: 30.3 (#77)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| CritPt | 0% | 10% |
| Chess Puzzles | 6% | 24% |
| LMArena Hard Prompts | 1438 | 1424 |
| DTBench | 74.9% | 80% |
| LMCA | 29.7% | 40.8% |
| Epoch Capabilities Index | 141.85 | 148.83 |
| ARC-AGI-2 | — | 18.9% |
| Kagi LLM Benchmark | — | 37.9% |
| NYT Connections (extended) | — | 61.8% |
| ARC-AGI-1 | — | 63.7% |
| Thematic Generalization | — | 61.7% |
| Mystery Game Puzzles | — | 11% |
| ForecastBench | — | 57 |
Math Too close to call
Gemma 4 26B A4B IT: 45.7 (#61), GPT-5.4 mini: 45.4 (#63)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 82.2% | 88.9% |
| LMArena Math | 1469 | 1418 |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 9.8% |
| ProofBench | — | 21% |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GPT-5.4 mini leads
Gemma 4 26B A4B IT: 45.6 (#79), GPT-5.4 mini: 51.4 (#62)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 73.2% | 86.9% |
| Vectara Hallucination Rate | 5.2% | 5.5% |
| LMArena Expert | 1447 | 1434 |
| SimpleQA Verified | — | 29.4% |
Multimodal Too close to call
Gemma 4 26B A4B IT: 39.9 (#43), GPT-5.4 mini: 39.0 (#51)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1260 | 1245 |
Multilingual Gemma 4 26B A4B IT leads
Gemma 4 26B A4B IT: 53.3 (#68), GPT-5.4 mini: 52.1 (#94)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1422 | 1405 |
| LMArena Chinese | 1495 | 1446 |
| LMArena French | 1459 | 1440 |
| LMArena Russian | 1434 | 1417 |
| LMArena Spanish | 1415 | 1403 |
| LMArena German | — | 1407 |
| LMArena Japanese | — | 1375 |
| LMArena Korean | — | 1369 |
Instruction Following Too close to call
Gemma 4 26B A4B IT: 75.0 (#78), GPT-5.4 mini: 74.3 (#99)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1420 | 1405 |
Long Context Too close to call
Gemma 4 26B A4B IT: 43.6 (#87), GPT-5.4 mini: 42.9 (#107)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1427 | 1407 |
Writing & Preference GPT-5.4 mini leads
Gemma 4 26B A4B IT: 58.8 (#110), GPT-5.4 mini: 64.2 (#56)
| Benchmark | Gemma 4 26B A4B IT | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1434 | 1411 |
| LMArena Creative Writing | 1401 | 1370 |
| EQ-Bench Creative Writing | 1305 | 1665 |
| LMArena Multi-Turn | 1442 | 1429 |
Frequently asked questions
Is Gemma 4 26B A4B IT better than GPT-5.4 mini?
GPT-5.4 mini is the stronger model overall, scoring 44.8 to 43.1 on the Noometry Index. Gemma 4 26B A4B IT costs 14× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 26B A4B IT or GPT-5.4 mini?
Gemma 4 26B A4B IT is cheaper. It lists at $0.0765 per million input tokens and $0.26 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is Gemma 4 26B A4B IT or GPT-5.4 mini better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 38.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 262K.
How many benchmarks do Gemma 4 26B A4B IT and GPT-5.4 mini share?
28 benchmarks have published results for both models. Gemma 4 26B A4B IT has 28 scored results on Noometry and GPT-5.4 mini has 46.