Model comparison
Gemma 3 4B vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 28.1 on the Noometry Index. Gemma 3 4B costs 7.8× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Google
Arcee AISummary
- They share 13 benchmarks with published results for both. Gemma 3 4B scores higher in 1 category and Trinity Large Thinking in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Trinity Large Thinking leads 40.9 to 11.8.
- Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- Trinity Large Thinking accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 3 4B | Trinity Large Thinking | |
|---|---|---|
| Provider | Arcee AI | |
| Noometry Index | 28.1 | 38.6 |
| Released | 2025-03-12 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 4K | 80K |
| Input $ / M tokens | $0.04 | $0.25 |
| Output $ / M tokens | $0.08 | $0.80 |
| Results tracked | 22 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 4B leads
Gemma 3 4B: 35.9 (#215), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Coding | 1230 | 1381 |
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |
Agentic & Tool Use Not comparable
Gemma 3 4B: 20.9 (#142), Trinity Large Thinking: —
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| Berkeley Function Calling Leaderboard | 19.6% | — |
Reasoning Trinity Large Thinking leads
Gemma 3 4B: 13.2 (#335), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Hard Prompts | 1253 | 1350 |
| Kagi LLM Benchmark | 25.2% | — |
| NYT Connections (extended) | — | 16.5% |
| CritPt | — | 0.9% |
| Chess Puzzles | 0% | — |
| Thematic Generalization | — | 41.6% |
| DTBench | 50.9% | — |
| LMCA | 2.8% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 116.02 | — |
Math Trinity Large Thinking leads
Gemma 3 4B: 16.8 (#292), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1239 | 1366 |
| OTIS Mock AIME 2024-2025 | 7.5% | — |
Knowledge Trinity Large Thinking leads
Gemma 3 4B: 11.8 (#299), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 6.4% | 6.9% |
| LMArena Expert | 1223 | 1360 |
| GPQA Diamond | 23.2% | — |
Multilingual Trinity Large Thinking leads
Gemma 3 4B: 42.5 (#194), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1273 | 1325 |
| LMArena German | 1281 | 1356 |
| LMArena Russian | 1294 | 1337 |
| LMArena Chinese | — | 1373 |
| LMArena French | — | 1374 |
| LMArena Japanese | — | 1311 |
| LMArena Korean | — | 1306 |
| LMArena Spanish | — | 1357 |
Instruction Following Trinity Large Thinking leads
Gemma 3 4B: 65.2 (#225), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1239 | 1334 |
Long Context Trinity Large Thinking leads
Gemma 3 4B: 38.7 (#194), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1273 | 1355 |
Writing & Preference Trinity Large Thinking leads
Gemma 3 4B: 42.0 (#239), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Gemma 3 4B | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1291 | 1340 |
| LMArena Creative Writing | 1271 | 1320 |
| LMArena Multi-Turn | 1255 | 1342 |
| EQ-Bench Creative Writing | 1068 | — |
Frequently asked questions
Is Gemma 3 4B better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 28.1 on the Noometry Index. Gemma 3 4B costs 7.8× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Which is cheaper, Gemma 3 4B or Trinity Large Thinking?
Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is Gemma 3 4B or Trinity Large Thinking better for coding?
Gemma 3 4B scores higher on coding benchmarks: 35.9 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 131K.
How many benchmarks do Gemma 3 4B and Trinity Large Thinking share?
13 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Trinity Large Thinking has 24.