Model comparison
Gemma 4 31B IT vs Trinity Large Thinking
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 38.1 on the Noometry Index.
Last verified . 21 shared benchmarks.
Google
Arcee AISummary
- They share 21 benchmarks with published results for both. Gemma 4 31B IT scores higher in 7 categories and Trinity Large Thinking in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 4 31B IT leads 26.4 to 16.3.
- The biggest single-benchmark swing is NYT Connections (extended): 70.6% for Gemma 4 31B IT and 16.5% for Trinity Large Thinking.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
Side by side
| Gemma 4 31B IT | Trinity Large Thinking | |
|---|---|---|
| Provider | Arcee AI | |
| Noometry Index | 43.2 | 38.1 |
| Released | 2026-04-02 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 33K | 80K |
| Input $ / M tokens | $0.09 | $0.25 |
| Output $ / M tokens | $0.34 | $0.80 |
| Results tracked | 35 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 41.9 (#98), Trinity Large Thinking: 33.8 (#238)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | 1364 | 1238 |
| SciCode | 43.4% | 36.1% |
| LMArena Coding | 1459 | 1381 |
| WeirdML | 52.3% | — |
| ALE-Bench | 925.5 | — |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 26.4 (#119), Trinity Large Thinking: 16.3 (#298)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | 70.6% | 16.5% |
| CritPt | 1.4% | 0.9% |
| Thematic Generalization | 53% | 41.6% |
| LMArena Hard Prompts | 1448 | 1350 |
| Surface Evolver Bench | 30.6% | 15.6% |
| Kagi LLM Benchmark | 63.5% | — |
| Chess Puzzles | 5% | — |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Epoch Capabilities Index | 142.74 | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 41.3 (#72), Trinity Large Thinking: 34.9 (#149)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1464 | 1366 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
Knowledge Trinity Large Thinking leads
Gemma 4 31B IT: 37.9 (#141), Trinity Large Thinking: 40.3 (#107)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 7.4% | 6.9% |
| LMArena Expert | 1465 | 1359 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
Multimodal Not comparable
Gemma 4 31B IT: 41.0 (#32), Trinity Large Thinking: —
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), Trinity Large Thinking: 46.4 (#155)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1431 | 1325 |
| LMArena Chinese | 1478 | 1372 |
| LMArena French | 1436 | 1374 |
| LMArena Russian | 1462 | 1338 |
| LMArena Spanish | 1443 | 1356 |
| LMArena German | — | 1356 |
| LMArena Japanese | — | 1312 |
| LMArena Korean | — | 1304 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.6 (#58), Trinity Large Thinking: 70.7 (#157)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1433 | 1334 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#67), Trinity Large Thinking: 41.2 (#139)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1446 | 1355 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.7 (#92), Trinity Large Thinking: 54.0 (#152)
| Benchmark | Gemma 4 31B IT | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1443 | 1340 |
| LMArena Creative Writing | 1416 | 1321 |
| LMArena Multi-Turn | 1452 | 1342 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Trinity Large Thinking?
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 38.1 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or Trinity Large Thinking?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is Gemma 4 31B IT or Trinity Large Thinking better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 41.9 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Gemma 4 31B IT and Trinity Large Thinking share?
21 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Trinity Large Thinking has 24.