Inference
EN
Arcee AI

Trinity-Large-Preview

Trinity-Large-Preview is a standard model from Arcee AI in the Trinity Large family. 10 benchmarks count toward its index, in 5 categories.

  • Open weights
  • Standard
  • Input: text
  • Released 2026-01
RankedEnough results for an index.
IndexAll categories, out of 100.42.2±7.6Median 51.0
CoverageIndex weight that has results.75%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.25Median $0.60
Output / 1MUS dollars per 1M output tokens.$1.00Median $2.55
ContextMaximum tokens in one request.512KMedian 262K
EloLMArena rating and rank.1339±4 #221Median 1422

Run locally

398.6B parameters
16-bit
838.1 GB
8-bit
445.7 GB
4-bit
254.7 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy needs more than 128 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.2 counted48.3145 / 218
CodingCode writing and repair.1 counted48.8—
ReasoningLogic problems and puzzles.2 counted36.0183 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.4 counted39.0198 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted41.6—
MathMath problems.None counted——

Results

10 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
τ²-Bench Tool-Agent-User EvaluationAgenticAbove fitFar above this model's index.90.1%40 / 149
Massive Multitask Language UnderstandingKnowledge87.2%12 / 34
Artificial Analysis IFBenchInstruction56.3%68 / 146
Artificial Analysis Humanity's Last ExamKnowledge15.8%123 / 222
Critical Physics TasksReasoning0.9%124 / 214
Artificial Analysis GPQA DiamondKnowledge75.2%116 / 199
Artificial Analysis SciCodeCoding40.6%84 / 114
GDPval-AA normalizedAgentic0.0%101 / 133
Artificial Analysis Long Context ReasoningReasoning38.0%169 / 217
GPQA Diamond · BenchLMKnowledge63.3%77 / 84
MMLU-Pro first-party comparison snapshotKnowledge75.2%6 / 6
AIME25 first-party comparison snapshotMath24.0%—

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index42.6 → 42.2
23 Oct 202612 Nov 2026
Output / 1M$1.00 → $1.00
23 Oct 202612 Nov 2026

Questions

How much does Trinity-Large-Preview cost?
$0.25 per 1M input tokens, $1.00 per 1M output tokens.
How good is Trinity-Large-Preview?
Its index is 42.2 out of 100. That is place 198 of 283 ranked models.
What is Trinity-Large-Preview best at?
Agentic: 48.3 out of 100, place 145 of 218.
What is the context window of Trinity-Large-Preview?
512K tokens.
Is Trinity-Large-Preview open source?
Yes. Its weights are published.
When was Trinity-Large-Preview released?
January 2026.

Badge

Trinity-Large-Preview on Inference360: place 198 of 283, index 42.2
HTML
<a href="https://inference360.ai/models/trinity-large-preview"><img src="https://inference360.ai/badges/trinity-large-preview.svg" alt="Trinity-Large-Preview on Inference360: place 198 of 283, index 42.2" height="20"></a>
Markdown
[![Trinity-Large-Preview on Inference360: place 198 of 283, index 42.2](https://inference360.ai/badges/trinity-large-preview.svg)](https://inference360.ai/models/trinity-large-preview)

Compare with

K-Exaone42.3GLM-4.641.8Trinity-Large-Thinking43.5

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from Arcee AI

All models
Trinity-Large-Thinking43.5