Gemma 3 12B gemma-3-12b-it
google · gemma efficient
#41 overall
- params
- 12B
- arch
- dense
- context
- 128k
- license
- gemma open weights gated
- released
- Mar 2025
- downloads/30d
- 1.3M
# why ranked
Overall: #41 — score 5.2 ● medium — one signal missing, weights renormalized
| signal | weight | input (0–100) |
|---|---|---|
| aa_intelligence | 0.50 | 5.5 |
| arena_elo | 0.30 | — |
| bench_composite | 0.20 | 4.6 |
missing signals are dropped and the remaining weights renormalized — never imputed.
Coding: provisional — score — ● 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 | 0.25 | 0.0 |
| aider_polyglot | 0.25 | — |
| arena_elo_code | 0.15 | — |
| swe_bench_verified | 0.35 | — |
missing signals are dropped and the remaining weights renormalized — never imputed.
# trend
# benchmarks
| benchmark | score | source | date |
|---|---|---|---|
| AA Coding Index | 5.8 | Artificial Analysis | — |
| AA Intelligence Index | 5.5 | Artificial Analysis | — |
| AA Math Index | 18.3 | Artificial Analysis | — |
| Ai2d | 0.8 / 1 | LLM Stats | — |
| Big-bench-extra-hard | 0.2 / 1 | LLM Stats | — |
| Big-bench-hard | 0.9 / 1 | LLM Stats | — |
| Bird-sql-(dev) | 0.5 / 1 | LLM Stats | — |
| Chartqa | 0.8 / 1 | LLM Stats | — |
| Docvqa | 0.9 / 1 | LLM Stats | — |
| Eclektic | 0.1 / 1 | LLM Stats | — |
# where to run
| provider | quant | ctx | $/M in | $/M out | $/M blend | tps | uptime |
|---|---|---|---|---|---|---|---|
| DeepInfra | bf16 | 128k | $0.05 | $0.15 | $0.08 | — | 100.0% |
blended = (3·input + output) / 4 per 1M tokens · cheapest first
source aliases
- aa
gemma-3-12b- arena
Gemma-3-12B-it- openrouter
google/gemma-3-12b-it