Intelligence
24 scoreArcee AI · Flagships Analysis
Trinity Large Thinking
Intelligence, Performance & Price Analysis
Canonical slug: trinity-large-thinking · Canonical model registry at build time
Speed
206.3 output tokens/secLatency
1.03s TTFTInput Price
$0.23 / 1M tokensOutput Price
$0.88 / 1M tokensExecutive Assessment
Routing Verdict & Tradeoffs
Ethen Routing Verdict
Strong general-purpose model
Trinity Large Thinking scores 24 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 25). Trinity Large Thinking generates output at 206.3 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight models of similar size (median: 61.7 t/s). Trinity Large Thinking costs $0.23 per 1M input tokens (very competitive, median: $0.59) and $0.88 per 1M output tokens (very competitive, median: $2.20), based on the median across providers serving the model.
Suitable for most production tasks, but high-volume or repetitive work should still be compared against cheaper routes.
Task Fit Assessment
| Workload | Rating | Notes |
|---|---|---|
| Complex reasoning & agentic workflows | optimal | Intelligence score 24 supports capable reasoning, but very hard tasks may benefit from higher-tier models. |
| High-volume chat & customer-facing | optimal | Output speed 206.3 tokens/sec and capable intelligence make this suitable for real-time chat at scale. |
| Latency-sensitive applications | optimal | TTFT 1.03s — adequate latency for most interactive use cases. |
| Cost-sensitive pipelines | optimal | Output pricing at $0.88 is reasonable for moderate volume. |
Cost Pressure Analysis
Low
Pricing is competitive — input $0.23, output $0.88. Suitable for sustained production use.
Technical Specifications
Architecture and Limits
| Specification | Value | Validation Authority |
|---|---|---|
| Model type | Open weights | inferred |
| Reasoning | Yes | faq |
| Input modalities | Trinity Large Thinking supports text input. | faq |
| Output modalities | Trinity Large Thinking supports text output. | faq |
| Context window | 510k tokens | faq |
| Open weights / source | Yes, Trinity Large Thinking is open weights. The model weights are publicly available and can be downloaded for self-hosting. | faq |
| Parameters | Trinity Large Thinking has 399 billion parameters (13 billion active). | faq |
| Active parameters | Trinity Large Thinking is a Mixture of Experts (MoE) model with 399 billion total parameters, but only 13 billion active parameters are used during inference. | faq |
| License | Trinity Large Thinking is released under the Apache 2.0 license. This license allows commercial use. | faq |
| API availability | Yes, Trinity Large Thinking is available via API through 2 providers. | faq |
Evidence charts
Profile visualizations
Charts are the same canonical evidence cards previously published for this slug, contained inside the D18D page grammar.
AA-Omniscience Index
AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Artificial Analysis Openness Index: Score
Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Intelligence
Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Output Speed
Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.
Speed
Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.
End-to-End Response Time
Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Latency: Time To First Answer Token
Seconds to first answer token received · Accounts for reasoning model 'thinking' time · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Latency includes thinking / reasoning time when applicable. Lower is better for interactive use cases.
Context Window
Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.
Cost per Intelligence Index Task
Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Cost per Task
Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Cost to Run Artificial Analysis Intelligence Index
Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.
Pricing: Cache Hit, Input, and Output
Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Bars intentionally show many models at once so the page feels like a benchmark report rather than a dashboard.
Time per Intelligence Index Task
Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Model Size: Total and Active Parameters
Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis
Retrieved 2026-07-08
Chart source and provenance are listed in Methodology & sources below.
- Source:
- Published benchmark dataset
- Retrieval date:
- 2026-07-08
- Methodology:
- Independent test run by Artificial Analysis on dedicated hardware.
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.
Methodology
Methodology & Provenance
This page is rendered from the normalized profile and page JSON for Trinity Large Thinking.
Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.
Frequently Asked Questions
Model FAQs & Technical Disclosures
Trinity Large Thinking was released on April 1, 2026.
Trinity Large Thinking was created by Arcee AI.
Trinity Large Thinking scores 24 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 25).
Trinity Large Thinking generates output at 206.3 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight models of similar size (median: 61.7 t/s).
Trinity Large Thinking has a time to first token (TTFT) of 1.03s (based on the median across providers serving the model), which is very competitive compared to other open weight models of similar size (median: 2.01s).
Trinity Large Thinking costs $0.23 per 1M input tokens (very competitive, median: $0.59) and $0.88 per 1M output tokens (very competitive, median: $2.20), based on the median across providers serving the model.
Trinity Large Thinking costs $0.23 per 1M input tokens and $0.88 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.24 per 1M tokens. Pricing may vary by provider.
Yes, Trinity Large Thinking is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Trinity Large Thinking supports text input.
Trinity Large Thinking supports text output.
No, Trinity Large Thinking does not support image input. It can only process text.
No, Trinity Large Thinking is not multimodal. It only supports text input.
Trinity Large Thinking has a context window of 510k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Trinity Large Thinking is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Trinity Large Thinking has 399 billion parameters (13 billion active).
Trinity Large Thinking is a Mixture of Experts (MoE) model with 399 billion total parameters, but only 13 billion active parameters are used during inference.
Trinity Large Thinking is released under the Apache 2.0 license. This license allows commercial use.
Trinity Large Thinking achieves a score of 24 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Trinity Large Thinking is available via API through 2 providers.
Trinity Large Thinking is available through 2 API providers.