Gemma 3n E4b gemma-3n-e4b
google · gemma
#108 overall
- params
- 7.8B
- license
- gemma open weights
# why ranked
Overall: #108 — score 2.3 ● low — a single signal; treat with caution
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.70 | 0.8 |
| bench_composite | 0.30 | 5.8 |
benchmark panel evidence:
| complete panel | score (0–100) |
|---|---|
| aa-coding-v1 | 11.6 |
| aa-math-v1 | 0.0 |
Coding: provisional — score 1.5 ● low — a single signal; treat with caution
provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.
| signal | weight | input (0–100) |
|---|---|---|
| aa_coding | 0.30 | 1.5 |
| aider_polyglot | 0.30 | — |
| swe_bench_verified | 0.40 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
methodology changed during this history window; score movement across that boundary is not model movement. See methodology v6.
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Coding Index | 3.2 | Artificial Analysis | — |
| AA Intelligence Index | 4.8 | Artificial Analysis | — |
| AA Math Index | 14.3 | Artificial Analysis | — |
| Arc-c | 0.6 / 1 | LLM Stats | — |
| Arc-e | 0.8 / 1 | LLM Stats | — |
| Big-bench-hard | 0.5 / 1 | LLM Stats | — |
| Boolq | 0.8 / 1 | LLM Stats | — |
# where to run
no known API providers — download the weights from Hugging Face and run them yourself.
source aliases
- aa
gemma-3n-e4b- llmstats
gemma-3n-e4b