Inference
EN
Arcee AI

Trinity-Large-Thinking

Trinity-Large-Thinking is a reasoning model from Arcee AI in the Trinity Large family. 12 benchmarks count toward its index, in 5 categories.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-03
RankedEnough results for an index.
IndexAll categories, out of 100.43.5±7.2Median 51.0
CoverageIndex weight that has results.75%
SpeedOutput tokens per second.138/sMedian 67/s
First tokenSeconds to the first output token.0.9 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.25Median $0.60
Output / 1MUS dollars per 1M output tokens.$0.80Median $2.55
ContextMaximum tokens in one request.262KMedian 262K
EloLMArena rating and rank.1340±5 #218Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
arcee-ai/trinity-large-thinking
Max outputMaximum output tokens per request.
80K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
No
Cache read / 1MUS dollars per 1M cached tokens.
$0.06
WeightsLink to the published weights.
Hugging Face (opens in a new tab) ↗
RetiresDate OpenRouter removes the model.
—

Run locally

398.64B parameters
16-bit
838.1 GB
8-bit
445.7 GB
4-bit
254.8 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.4 counted44.0165 / 218
CodingCode writing and repair.2 counted48.3149 / 209
ReasoningLogic problems and puzzles.2 counted36.0183 / 227
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted45.5166 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted39.8—
MathMath problems.None counted——

Results

12 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
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
MMLU-Pro first-party comparison snapshotKnowledge83.4%4 / 6
Artificial Analysis SciCodeCoding40.6%84 / 114
GDPval-AA normalizedAgentic0.0%101 / 133
Artificial Analysis Long Context ReasoningReasoning38.0%169 / 217
GPQA Diamond · BenchLMKnowledge76.3%67 / 84
Terminal-Bench 2.0 · Vals AIAgentic18.0% ±4.159 / 66
Instruction Following BenchmarkInstruction52.3%51 / 57
Gert Labs Composite Game BenchmarkAgentic32.5%47 / 52
SWE-bench Verified (mini-swe-agent-v2)Coding63.2%5 / 5
CaseLaw v2Legal57.9% ±1.929 / 53
CorpFin v2Finance54.7% ±1.092 / 131
AIME25 first-party comparison snapshotMath96.3%—
τ²-Bench Airline DomainAgentic88.0%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseVals AINo licence stated

History

5 releases
Index43.9 → 43.5
23 Oct 202612 Nov 2026
Output / 1M$0.80 → $0.80
23 Oct 202612 Nov 2026

Questions

How much does Trinity-Large-Thinking cost?
$0.25 per 1M input tokens, $0.80 per 1M output tokens.
How good is Trinity-Large-Thinking?
Its index is 43.5 out of 100. That is place 187 of 283 ranked models.
What is Trinity-Large-Thinking best at?
Knowledge: 45.5 out of 100, place 166 of 268.
How fast is Trinity-Large-Thinking?
138 output tokens per second, 0.9 s to the first token.
What is the context window of Trinity-Large-Thinking?
262K tokens.
Is Trinity-Large-Thinking open source?
Yes. Its weights are published.
When was Trinity-Large-Thinking released?
March 2026.

Badge

Trinity-Large-Thinking on Inference360: place 187 of 283, index 43.5
HTML
<a href="https://inference360.ai/models/trinity-large-thinking"><img src="https://inference360.ai/badges/trinity-large-thinking.svg" alt="Trinity-Large-Thinking on Inference360: place 187 of 283, index 43.5" height="20"></a>
Markdown
[![Trinity-Large-Thinking on Inference360: place 187 of 283, index 43.5](https://inference360.ai/badges/trinity-large-thinking.svg)](https://inference360.ai/models/trinity-large-thinking)

Compare with

Claude Haiku 4.5 Thinking43.5MiMo-V2-Flash43.4GPT-OSS 120B43.4Trinity-Large-Preview42.2

Lower price

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

More from Arcee AI

All models
Trinity-Large-Preview42.2