Model comparison
Qwen Turbo vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 27.1 on the Noometry Index. Qwen Turbo costs 4.4× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Alibaba (Qwen)
Arcee AISummary
- The widest gap is in math, where Trinity Large Thinking leads 37.6 to 15.3.
- Qwen Turbo is cheaper at $0.05 / $0.20 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- Qwen Turbo accepts more context: 1M tokens versus 262K.
- Trinity Large Thinking has downloadable open weights; the other is API-only.
Side by side
| Qwen Turbo | Trinity Large Thinking | |
|---|---|---|
| Provider | Alibaba (Qwen) | Arcee AI |
| Noometry Index | 27.1 | 38.6 |
| Released | 2024-11-01 | 2026-04-01 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 16K | 80K |
| Input $ / M tokens | $0.05 | $0.25 |
| Output $ / M tokens | $0.20 | $0.80 |
| Results tracked | 3 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Qwen Turbo: —, Trinity Large Thinking: 34.1 (#244)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |
| LMArena Coding | — | 1381 |
Reasoning Not comparable
Qwen Turbo: —, Trinity Large Thinking: 16.9 (#298)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | — | 16.5% |
| CritPt | — | 0.9% |
| Thematic Generalization | — | 41.6% |
| LMArena Hard Prompts | — | 1350 |
| Surface Evolver Bench | — | 15.6% |
Math Trinity Large Thinking leads
Qwen Turbo: 15.3 (#297), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.1% | — |
| LMArena Math | — | 1366 |
| MATH Level 5 | 56.2% | — |
Knowledge Trinity Large Thinking leads
Qwen Turbo: 22.2 (#272), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| GPQA Diamond | 41.8% | — |
| Vectara Hallucination Rate | — | 6.9% |
| LMArena Expert | — | 1360 |
Multilingual Not comparable
Qwen Turbo: —, Trinity Large Thinking: 46.2 (#160)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | — | 1325 |
| LMArena Chinese | — | 1373 |
| LMArena French | — | 1374 |
| LMArena German | — | 1356 |
| LMArena Japanese | — | 1311 |
| LMArena Korean | — | 1306 |
| LMArena Russian | — | 1337 |
| LMArena Spanish | — | 1357 |
Instruction Following Not comparable
Qwen Turbo: —, Trinity Large Thinking: 70.5 (#162)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | — | 1334 |
Long Context Not comparable
Qwen Turbo: —, Trinity Large Thinking: 41.3 (#144)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | — | 1355 |
Writing & Preference Not comparable
Qwen Turbo: —, Trinity Large Thinking: 53.8 (#158)
| Benchmark | Qwen Turbo | Trinity Large Thinking |
|---|---|---|
| LMArena Text | — | 1340 |
| LMArena Creative Writing | — | 1320 |
| LMArena Multi-Turn | — | 1342 |
Frequently asked questions
Is Qwen Turbo better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 27.1 on the Noometry Index. Qwen Turbo costs 4.4× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Which is cheaper, Qwen Turbo or Trinity Large Thinking?
Qwen Turbo is cheaper. It lists at $0.05 per million input tokens and $0.20 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Which has the bigger context window?
Qwen Turbo does, with 1M tokens against 262K.
How many benchmarks do Qwen Turbo and Trinity Large Thinking share?
0 benchmarks have published results for both models. Qwen Turbo has 3 scored results on Noometry and Trinity Large Thinking has 24.