Google
availableShows if the model has enough results for an index.Gemma 4 26B A4B
Gemma 4 26B A4B is a reasoning model from Google in the Gemma 4 family. 19 benchmarks count toward its score, in 8 categories.
IndexOverall score out of 100.45.2 ±3.7
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.—
Input / 1MUS dollars per 1M input tokens.Free
Output / 1MUS dollars per 1M output tokens.Free
ContextMaximum tokens in one request.256K
EloLMArena rating and rank.1434 (#80)
The index is a score out of 100. The ± range shows how much it can change. A free tier is available.
5,804 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category, out of 100.
Out of 100Results
19 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| Massive Multitask Language Understanding Professional | Knowledge | 82.6% | 50.7 | — | — | Yubo Wang et al. |
| OTIS Mock AIME 2024-2025 | Math | 82.2% | 57.5 | minimal effort | — | Epoch AI |
| Artificial Analysis GPQA Diamond | Knowledge | 79.2% | 49.7 | — | — | Artificial Analysis |
| Massive Multi-discipline Multimodal Understanding Pro | Multimodal | 73.8% | 47.4 | — | — | MMMU-Pro authors |
| GPQA diamond | Knowledge | 73.2% | 45.9 | minimal effort | — | Epoch AI |
| Artificial Analysis IFBench | Instruction | 72.4% | 62.6 | — | — | Artificial Analysis |
| Artificial Analysis MMMU-Pro | Multimodal | 69.2% | 51.7 | — | — | Artificial Analysis |
| Artificial Analysis Long Context Reasoning | Reasoning | 65.7% | 53.7 | — | — | Artificial Analysis |
| EuroEval French | Multilingual | 65.2% | 83.7 | — | — | EuroEval |
| EuroEval Portuguese | Multilingual | 62.0% | 79.6 | — | — | EuroEval |
| EuroEval Italian | Multilingual | 61.1% | 78.6 | — | — | EuroEval |
| EuroEval Swedish | Multilingual | 58.8% | 75.7 | — | — | EuroEval |
| EuroEval Dutch | Multilingual | 56.4% | 72.7 | — | — | EuroEval |
| EuroEval Polish | Multilingual | 55.5% | 71.6 | — | — | EuroEval |
| EuroEval Spanish | Multilingual | 49.4% | 64.0 | — | — | EuroEval |
| EuroEval German | Multilingual | 49.1% | 63.7 | — | — | EuroEval |
| EuroEval Dutch | Multilingual | 45.3% | 58.9 | — | — | EuroEval |
| EuroEval French | Multilingual | 44.4% | 57.8 | — | — | EuroEval |
| EuroEval Swedish | Multilingual | 44.4% | 57.8 | — | — | EuroEval |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 43.6% | 30.1 | — | — | Victor Barres et al. |
| EuroEval Portuguese | Multilingual | 43.5% | 56.7 | — | — | EuroEval |
| EuroEval Italian | Multilingual | 41.0% | 53.5 | — | — | EuroEval |
| Artificial Analysis SciCode | Coding | 40.0% | 48.4 | — | — | Artificial Analysis |
| Artificial Analysis Coding Index | Coding | 39.3% | 46.7 | — | — | Artificial Analysis |
| EuroEval Polish | Multilingual | 38.5% | 50.4 | — | — | EuroEval |
| EuroEval German | Multilingual | 37.7% | 49.4 | — | — | EuroEval |
| EuroEval Spanish | Multilingual | 35.4% | 46.5 | — | — | EuroEval |
| Artificial Analysis Humanity's Last Exam | Knowledge | 19.3% | 45.5 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 19.1% | 37.5 | — | — | Artificial Analysis |
| Humanity's Last Exam | Knowledge | 17.2% | 43.4 | — | — | Center for AI Safety et al. |
| Artificial Analysis Intelligence Index | Knowledge | 16.7% | 43.2 | — | — | Artificial Analysis |
| Humanity's Last Exam without tools | Knowledge | 8.7% | 36.2 | — | — | OpenAI |
| Chess Puzzles | Reasoning | 6.0% | 31.3 | minimal effort | — | Epoch AI |
| GDPval-AA normalized | Agentic | 2.6% | 37.6 | — | — | Artificial Analysis |
| Critical Physics Tasks | Reasoning | 0.0% | 38.4 | — | — | Artificial Analysis |
19 benchmarks count, from 35 of 35 results. A grey row does not count. Too few models took that benchmark.