Gemma 4 31Bopen
36.8#289 of 427 overall
- Input / output
- US$ 0,20 / US$ 0,40 US$ per 1M tokens
- Context
- 262K tokens
- Release
- 04/02/2026 release date
- Speed
- 35 tok/s first token in 0.95 s
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| LiveBench weight 3 | 38.7 | 0% | LiveBench |
| Arena text (Elo) weight 2 | 1443 | 73% | LMArena |
| Arena WebDev (Elo) weight 1 | 1369 | 24% | LMArena |
| Arena vision (Elo) weight 0.5 | 1276 | 71% | LMArena |
| Humanity's Last Exam weight 1 | 23.6 | 72% | Artificial Analysis |
| Terminal-Bench weight 1 | 0.0 | 22% | Artificial Analysis |
| GPQA Diamond | 85.7 | 80% | Artificial Analysis |
| Long-context reasoning | 69.7 | 61% | Artificial Analysis |
| LiveBench long agentic | 10.7 | 13% | LiveBench |
| LiveBench reasoning | 68.3 | 0% | LiveBench |
| LiveBench coding | 49.0 | 0% | LiveBench |
| LiveBench browser | 0.0 | 0% | LiveBench |
| LiveBench data analysis | 53.3 | 3% | LiveBench |
| LiveBench research | 44.4 | 7% | LiveBench |
| LiveBench design | 58.4 | 7% | LiveBench |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 28.3 | #145 | LiveBench long agentic, Terminal-Bench |
| Coding | 24.7 | #120 | LiveBench coding, Arena WebDev (Elo) |
| Reasoning | 50.5 | #212 | LiveBench reasoning, Humanity's Last Exam, GPQA Diamond |
| Data analysis | 26.6 | #31 | LiveBench data analysis |
| Research | 39.1 | #256 | LiveBench research, Long-context reasoning |
| Design | 26.9 | #114 | LiveBench design, Arena WebDev (Elo) |
| Browser use | 25.0 | #32 | LiveBench browser |
| Chat | 61.4 | #56 | Arena text (Elo) |
| Vision | 60.3 | #32 | Arena vision (Elo) |
History
RankingIA score on each collection day (hover to see the rank).
Daily collection started on 10/30/2026. The chart appears from the second day the model is ranked.