Model comparison
GPT-5 Nano vs Trinity Large Thinking
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 33.5 on the Noometry Index. GPT-5 Nano costs 2.8× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Arcee AISummary
- They share 17 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Trinity Large Thinking in 7 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Trinity Large Thinking leads 53.8 to 39.1.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $0.25 / $0.80 for Trinity Large Thinking.
- GPT-5 Nano accepts more context: 400K tokens versus 262K.
- Trinity Large Thinking has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Trinity Large Thinking | |
|---|---|---|
| Provider | OpenAI | Arcee AI |
| Noometry Index | 33.5 | 38.6 |
| Released | 2025-08-07 | 2026-04-01 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 80K |
| Input $ / M tokens | $0.05 | $0.25 |
| Output $ / M tokens | $0.40 | $0.80 |
| Results tracked | 49 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5 Nano: 33.6 (#254), Trinity Large Thinking: 34.1 (#244)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Coding | 1351 | 1381 |
| SWE-bench Verified (bash only) | 34.8% | — |
| LMArena WebDev | — | 1238 |
| SciCode | — | 36.1% |
| WeirdML | 38.1% | — |
| ALE-Bench | 718.67 | — |
Agentic & Tool Use Not comparable
GPT-5 Nano: 25.8 (#106), Trinity Large Thinking: —
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
Reasoning Too close to call
GPT-5 Nano: 16.3 (#306), Trinity Large Thinking: 16.9 (#298)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Hard Prompts | 1328 | 1350 |
| ARC-AGI-2 | 2.6% | — |
| Kagi LLM Benchmark | 62.2% | — |
| NYT Connections (extended) | — | 16.5% |
| ARC-AGI-1 | 20.7% | — |
| CritPt | — | 0.9% |
| Chess Puzzles | 27% | — |
| Thematic Generalization | — | 41.6% |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 139.38 | — |
| ForecastBench | 59.1 | — |
Math Trinity Large Thinking leads
GPT-5 Nano: 29.4 (#241), Trinity Large Thinking: 37.6 (#149)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1317 | 1366 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Trinity Large Thinking leads
GPT-5 Nano: 35.9 (#178), Trinity Large Thinking: 40.9 (#113)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 10.5% | 6.9% |
| LMArena Expert | 1321 | 1360 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| GPQA (HELM) | 67.9% | — |
Multimodal Not comparable
GPT-5 Nano: 31.3 (#108), Trinity Large Thinking: —
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual Too close to call
GPT-5 Nano: 45.3 (#172), Trinity Large Thinking: 46.2 (#160)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1313 | 1325 |
| LMArena Chinese | 1356 | 1373 |
| LMArena German | 1327 | 1356 |
| LMArena Japanese | 1226 | 1311 |
| LMArena Korean | 1269 | 1306 |
| LMArena Russian | 1296 | 1337 |
| LMArena Spanish | 1360 | 1357 |
| LMArena French | — | 1374 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), Trinity Large Thinking: 70.5 (#162)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1306 | 1334 |
| IFEval | 93.2% | — |
Long Context Trinity Large Thinking leads
GPT-5 Nano: 31.3 (#281), Trinity Large Thinking: 41.3 (#144)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1312 | 1355 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Trinity Large Thinking leads
GPT-5 Nano: 39.1 (#249), Trinity Large Thinking: 53.8 (#158)
| Benchmark | GPT-5 Nano | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1320 | 1340 |
| LMArena Creative Writing | 1249 | 1320 |
| LMArena Multi-Turn | 1311 | 1342 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
Frequently asked questions
Is GPT-5 Nano better than Trinity Large Thinking?
Trinity Large Thinking is the stronger model overall, scoring 38.6 to 33.5 on the Noometry Index. GPT-5 Nano costs 2.8× less per token, which makes it the better buy when Trinity Large Thinking's lead doesn't matter for your workload.
Which is cheaper, GPT-5 Nano or Trinity Large Thinking?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; Trinity Large Thinking lists at $0.25 and $0.80.
Is GPT-5 Nano or Trinity Large Thinking better for coding?
They score almost the same on coding (33.6 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5 Nano does, with 400K tokens against 262K.
How many benchmarks do GPT-5 Nano and Trinity Large Thinking share?
17 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Trinity Large Thinking has 24.