Trinity Models by Arcee AI
Arcee AIApache 2.0
4 models2025–2026Up to 256K ctxFrom $0.045/1M input
About
Arcee AI's open-source large language model family.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
4 in view
Trinity-Large-ThinkingCurrent
Use when the workload needs reasoning, 256K context, and 400B parameters.
2026-04reasoning256K context400B parameters
Trinity-Large-PreviewCurrent
Use when the workload needs 128K context, 400B parameters, and tool use.
2026-01128K context400B parameterstool use
Trinity MiniCurrent
Use when the workload needs 128K context, 26B parameters, and tool use.
2025-12128K context26B parameterstool use
Trinity NanoCurrent
Use when the workload needs 128K context and 6B parameters.
2025-12128K context6B parameters
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Trinity-Large-Thinking | Use when the workload needs reasoning, 256K context, and 400B parameters. | 2026-04 | reasoning256K context400B parameters | Current |
| Trinity-Large-Preview | Use when the workload needs 128K context, 400B parameters, and tool use. | 2026-01 | 128K context400B parameterstool use | Current |
| Trinity Mini | Use when the workload needs 128K context, 26B parameters, and tool use. | 2025-12 | 128K context26B parameterstool use | Current |
| Trinity Nano | Use when the workload needs 128K context and 6B parameters. | 2025-12 | 128K context6B parameters | Current |
Release Timeline
3 release groups2026-04
1 current
Trinity-Large-Thinking
Currentreasoning256K context400B parameters
2026-01
1 current
Trinity-Large-Preview
Current128K context400B parameterstool use
2025-12
2 current
Trinity Mini
Current128K context26B parameterstool use
Trinity Nano
Current128K context6B parameters
Specifications(4 models)
| Model | Released | Context | Parameters | Reasoning | Fn Calling | Tool Use | Structured Outputs |
|---|---|---|---|---|---|---|---|
| Trinity-Large-Thinking | 2026-04 | 256K | 400B | Yes | Yes | Yes | Yes |
| Trinity-Large-Preview | 2026-01 | 128K | 400B | No | Yes | Yes | Yes |
| Trinity Mini | 2025-12 | 128K | 26B | No | Yes | Yes | Yes |
| Trinity Nano | 2025-12 | 128K | 6B | No | No | No | No |
Available From(3 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Trinity Mini | OpenRouter | $0.045 | $0.15 | Serverless |
| Trinity Mini | Vercel AI Gateway | $0.045 | $0.15 | Serverless |
| Trinity-Large-Preview | OpenRouter | $0.15 | $0.45 | Serverless |
| Trinity-Large-Thinking | OpenRouter | $0.22 | $0.85 | Serverless |
| Trinity-Large-Preview | Vercel AI Gateway | $0.25 | $1 | Serverless |
| Trinity-Large-Thinking | Vercel AI Gateway | $0.25 | $0.9 | Serverless |
Frequently Asked Questions
- What is Trinity used for?
- Trinity is used for reasoning, agent workflows and tool use, and structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
- How does Trinity compare to AFM?
- Trinity by Arcee AI is strongest where you need reasoning, while AFM by Arcee AI is the closest related family to check for adjacent model selection. Trinity has 4 listed variants and reaches up to 256K context, while AFM reaches up to 66K context, so compare the specs and pricing tables before choosing a production model.
- Which Trinity model should I use?
- For the lowest listed input price, start with Trinity Mini through OpenRouter at $0.045/1M input tokens. For the most capable/latest local choice, evaluate Trinity-Large-Thinking with 256K context and reasoning, tool use, function calling, and structured outputs.
