Arcee AI Trinity Mini

Advanced sparse MoE model for efficient inference on Together AI.

About model

Trinity Mini brings frontier-level language understanding to your applications without frontier costs. This 26B sparse mixture-of-experts model activates just 3B parameters per token, delivering exceptional reasoning, tool use, and multi-turn conversation capabilities across a 128K context window. Whether you're building conversational AI, agentic workflows, or production systems requiring long-context understanding, Trinity Mini offers the efficiency and performance to scale from prototype to production seamlessly.

Performance benchmarks

Model	GPQA Diamond	HLE	LiveCodeBench	MATH500	SWE-bench verified
Arcee AI Trinity Mini	58.6%			92.1%
Related open-source models
Competitor closed-source models
Claude Opus 4.6	90.5%	34.2%			78.7%
OpenAI o3	83.3%	24.9%		99.2%	62.3%
OpenAI o1	76.8%			96.4%	48.9%
GPT-4o	49.2%	2.7%	32.3%	89.3%	31.0%

API usage

cURL
Python
Typescript

Endpoint:

arcee-ai/trinity-mini

curl -X POST "https://api.together.xyz/v1/chat/completions" \
  -H "Authorization: Bearer $TOGETHER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "arcee-ai/trinity-mini",
    "messages": [
      {
        "role": "user",
        "content": "What are some fun things to do in New York?"
      }
    ]
}'

from together import Together

client = Together()

response = client.chat.completions.create(
  model="arcee-ai/trinity-mini",
  messages=[
    {
      "role": "user",
      "content": "What are some fun things to do in New York?"
    }
  ]
)
print(response.choices[0].message.content)

import Together from 'together-ai';
const together = new Together();

const completion = await together.chat.completions.create({
  model: 'arcee-ai/trinity-mini',
  messages: [
    {
      role: 'user',
      content: 'What are some fun things to do in New York?'
     }
  ],
});

console.log(completion.choices[0].message.content);

Model card
Architecture Overview:
• Sparse mixture-of-experts (MoE) architecture with 26B total parameters and 3B activated per token
• Efficient attention mechanism that reduces memory and compute requirements while preserving long-context coherence
• 128K-token context window supporting extended document processing and multi-turn interactions

Training Methodology:
• Trained with continuous reinforcement learning techniques for ongoing capability improvements
• Built by Arcee AI's collaborative research team focused on delivering best-in-class per-parameter performance
• Optimized for multi-turn conversations, tool use, and structured outputs

Performance Characteristics:
• Strong context utilization that fully leverages long inputs for coherent multi-turn reasoning
• Reliable function and tool calling capabilities for agent workflows
• High inference efficiency generating tokens rapidly while minimizing compute
• Outstanding price-to-performance ratio compared to dense models of similar capability
‍
Applications & use cases
Conversational AI Applications:
• Multi-turn customer support chatbots with long conversation history
• Virtual assistants with tool integration and function calling
• Interactive documentation and knowledge base systems

Agentic Workflows:
• Multi-step agent systems requiring tool use and reasoning
• Workflow automation with structured output generation
• RAG systems with extended context understanding

Enterprise Integration:
• Cost-efficient production deployments via Together AI APIs
• Internal tooling with natural language interfaces
• Document analysis and processing pipelines with 128K context support
‍

Related models

Model specifications

Model data

Model provider
Arcee AI
Type
Chat
Main use cases
Chat
Small & Fast
Deployment
Serverless
Monthly Reserved
Endpoint
arcee-ai/trinity-mini
Parameters
26B
Activated parameters
3B
Context length
32.7K
Input price
$0.05 / 1M tokens
Output price
$0.15 / 1M tokens
Input modalities
Text
Output modalities
Text

Released
December 1, 2025
External link
Provider docs
Category
Chat

Run in Playground

Quickstart docs

Deploy model

Arcee AI Trinity Mini

About model

API usage

Model card

Applications & use cases