Model comparison
GPT-5.4 mini vs Trinity Large Thinking
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 38.6 on the Noometry Index. Trinity Large Thinking costs 4.4× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Arcee AISummary
- They share 23 benchmarks with published results for both. GPT-5.4 mini scores higher in 8 categories and Trinity Large Thinking in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 mini leads 30.4 to 16.9.
- The biggest single-benchmark swing is NYT Connections (extended): 61.8% for GPT-5.4 mini and 16.5% for Trinity Large Thinking.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 262K.
- Trinity Large Thinking has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | Trinity Large Thinking | |
|---|---|---|
| Provider | OpenAI | Arcee AI |
| Noometry Index | 45.0 | 38.6 |
| Released | 2026-03-17 | 2026-04-01 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 80K |
| Input $ / M tokens | $0.75 | $0.25 |
| Output $ / M tokens | $4.50 | $0.80 |
| Results tracked | 46 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.2 (#72), Trinity Large Thinking: 34.1 (#244)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | 1397 | 1238 |
| SciCode | 49.9% | 36.1% |
| LMArena Coding | 1438 | 1381 |
| FrontierCode | 27% | — |
| WeirdML | 60.3% | — |
| ALE-Bench | 1,189 | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.9 (#81), Trinity Large Thinking: —
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.4 (#85), Trinity Large Thinking: 16.9 (#298)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | 61.8% | 16.5% |
| CritPt | 10% | 0.9% |
| Thematic Generalization | 61.7% | 41.6% |
| LMArena Hard Prompts | 1424 | 1350 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | 24% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 148.84 | — |
| ForecastBench | 57 | — |
Math GPT-5.4 mini leads
GPT-5.4 mini: 45.5 (#75), Trinity Large Thinking: 37.6 (#149)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1419 | 1366 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 21% | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.5 (#67), Trinity Large Thinking: 40.9 (#113)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 6.9% |
| LMArena Expert | 1435 | 1360 |
| GPQA Diamond | 86.9% | — |
| SimpleQA Verified | 29.4% | — |
Multimodal Not comparable
GPT-5.4 mini: 39.7 (#56), Trinity Large Thinking: —
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual GPT-5.4 mini leads
GPT-5.4 mini: 51.9 (#96), Trinity Large Thinking: 46.2 (#160)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1405 | 1325 |
| LMArena Chinese | 1446 | 1373 |
| LMArena French | 1440 | 1374 |
| LMArena German | 1409 | 1356 |
| LMArena Japanese | 1374 | 1311 |
| LMArena Korean | 1368 | 1306 |
| LMArena Russian | 1417 | 1337 |
| LMArena Spanish | 1405 | 1357 |
Instruction Following GPT-5.4 mini leads
GPT-5.4 mini: 74.1 (#102), Trinity Large Thinking: 70.5 (#162)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1405 | 1334 |
Long Context GPT-5.4 mini leads
GPT-5.4 mini: 43.0 (#112), Trinity Large Thinking: 41.3 (#144)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1407 | 1355 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.0 (#58), Trinity Large Thinking: 53.8 (#158)
| Benchmark | GPT-5.4 mini | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1412 | 1340 |
| LMArena Creative Writing | 1370 | 1320 |
| LMArena Multi-Turn | 1429 | 1342 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than Trinity Large Thinking?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 38.6 on the Noometry Index. Trinity Large Thinking costs 4.4× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini 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; GPT-5.4 mini lists at $0.75 and $4.50.
Is GPT-5.4 mini or Trinity Large Thinking better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 262K.
How many benchmarks do GPT-5.4 mini and Trinity Large Thinking share?
23 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Trinity Large Thinking has 24.