Model comparison
gpt-oss-120b vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 36.3 on the Noometry Index. gpt-oss-120b costs 5.5× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Arcee AISummary
- They share 21 benchmarks with published results for both. gpt-oss-120b scores higher in 4 categories and Trinity Large Thinking in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 37.6.
- The biggest single-benchmark swing is Surface Evolver Bench: 25% for gpt-oss-120b and 15.6% for Trinity Large Thinking.
- gpt-oss-120b is cheaper at $0.037 / $0.17 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
| gpt-oss-120b | Trinity Large Thinking | |
|---|---|---|
| Provider | OpenAI | Arcee AI |
| Noometry Index | 36.3 | 38.6 |
| Released | 2025-08-05 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 41K | 80K |
| Input $ / M tokens | $0.037 | $0.25 |
| Output $ / M tokens | $0.17 | $0.80 |
| Results tracked | 48 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
gpt-oss-120b: 33.5 (#256), Trinity Large Thinking: 34.1 (#244)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| SciCode | 36% | 36.1% |
| LMArena Coding | 1380 | 1381 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1238 |
| WeirdML | 48.2% | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Not comparable
gpt-oss-120b: 12.2 (#153), Trinity Large Thinking: —
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning gpt-oss-120b leads
gpt-oss-120b: 20.0 (#245), Trinity Large Thinking: 16.9 (#298)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| CritPt | 1.1% | 0.9% |
| LMArena Hard Prompts | 1364 | 1350 |
| Surface Evolver Bench | 25% | 15.6% |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 16.5% |
| Chess Puzzles | 20% | — |
| Thematic Generalization | — | 41.6% |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Epoch Capabilities Index | 139.93 | — |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Trinity Large Thinking: 37.6 (#149)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1389 | 1366 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Trinity Large Thinking: 40.9 (#113)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 14.2% | 6.9% |
| LMArena Expert | 1356 | 1360 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| GPQA (HELM) | 68.4% | — |
Multilingual gpt-oss-120b leads
gpt-oss-120b: 48.0 (#147), Trinity Large Thinking: 46.2 (#160)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1351 | 1325 |
| LMArena Chinese | 1385 | 1373 |
| LMArena French | 1369 | 1374 |
| LMArena German | 1353 | 1356 |
| LMArena Japanese | 1331 | 1311 |
| LMArena Korean | 1282 | 1306 |
| LMArena Russian | 1343 | 1337 |
| LMArena Spanish | 1389 | 1357 |
Instruction Following Trinity Large Thinking leads
gpt-oss-120b: 69.3 (#173), Trinity Large Thinking: 70.5 (#162)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1318 | 1334 |
| IFEval | 83.6% | — |
Long Context Trinity Large Thinking leads
gpt-oss-120b: 31.4 (#278), Trinity Large Thinking: 41.3 (#144)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1319 | 1355 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Trinity Large Thinking leads
gpt-oss-120b: 46.5 (#217), Trinity Large Thinking: 53.8 (#158)
| Benchmark | gpt-oss-120b | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1365 | 1340 |
| LMArena Creative Writing | 1275 | 1320 |
| LMArena Multi-Turn | 1340 | 1342 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 36.3 on the Noometry Index. gpt-oss-120b costs 5.5× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or Trinity Large Thinking?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is gpt-oss-120b or Trinity Large Thinking better for coding?
They score almost the same on coding (33.5 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 131K.
How many benchmarks do gpt-oss-120b and Trinity Large Thinking share?
21 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Trinity Large Thinking has 24.