Model comparison
DeepSeek-V3.1-Terminus vs Trinity Large Thinking
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 38.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
DeepSeek
Arcee AISummary
- They share 12 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Trinity Large Thinking in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1-Terminus leads 26.4 to 16.9.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $0.27 / $1 for DeepSeek-V3.1-Terminus.
- Trinity Large Thinking accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1-Terminus | Trinity Large Thinking | |
|---|---|---|
| Provider | DeepSeek | Arcee AI |
| Noometry Index | 43.1 | 38.6 |
| Released | 2025-09-22 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 147K | 80K |
| Input $ / M tokens | $0.27 | $0.25 |
| Output $ / M tokens | $1 | $0.80 |
| Results tracked | 16 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Trinity Large Thinking: 34.1 (#244)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| SciCode | 40.6% | 36.1% |
| LMArena Coding | 1426 | 1381 |
| LMArena WebDev | — | 1238 |
| ALE-Bench | 745.17 | — |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Trinity Large Thinking: 16.9 (#298)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| CritPt | 1.7% | 0.9% |
| LMArena Hard Prompts | 1426 | 1350 |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 16.5% |
| Thematic Generalization | — | 41.6% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Surface Evolver Bench | — | 15.6% |
Math Too close to call
DeepSeek-V3.1-Terminus: 38.5 (#137), Trinity Large Thinking: 37.6 (#149)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1402 | 1366 |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Trinity Large Thinking: 40.9 (#113)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | — | 6.9% |
| LMArena Expert | — | 1360 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Trinity Large Thinking: 46.2 (#160)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1407 | 1325 |
| LMArena Russian | 1436 | 1337 |
| LMArena Chinese | — | 1373 |
| LMArena French | — | 1374 |
| LMArena German | — | 1356 |
| LMArena Japanese | — | 1311 |
| LMArena Korean | — | 1306 |
| LMArena Spanish | — | 1357 |
Instruction Following DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Trinity Large Thinking: 70.5 (#162)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1404 | 1334 |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Trinity Large Thinking: 41.3 (#144)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1421 | 1355 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Trinity Large Thinking: 53.8 (#158)
| Benchmark | DeepSeek-V3.1-Terminus | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1419 | 1340 |
| LMArena Creative Writing | 1403 | 1320 |
| LMArena Multi-Turn | 1411 | 1342 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Trinity Large Thinking?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 38.6 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1-Terminus or Trinity Large Thinking?
Trinity Large Thinking is cheaper. It lists at $0.25 per million input tokens and $0.80 per million output tokens; DeepSeek-V3.1-Terminus lists at $0.27 and $1.
Is DeepSeek-V3.1-Terminus or Trinity Large Thinking better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and Trinity Large Thinking share?
12 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Trinity Large Thinking has 24.