MODEL DIRECTORY
阿西人工智能 Model Directory
Arcee AI is an artificial intelligence company focused on developing and deploying efficient open-source large language models.
Public models
4
With relay prices
0
Relay sites
0
Reasoning
2
Tool call
4
Structured output
0
Multimodal input
0
Text models
4
Image models
0
Video models
0
Audio models
0
Release range
2025-12-01 ~ 2026-04-01
4 models foundPrices in USD per 1M tokens; relay prices and ≈ are converted from CNY at today's rate
| Model | Provider | Type | Relays | Context | Max output | Input | Output | Capabilities | Price (input / output) | Details | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Trinity Large Thinkingtrinity-large-thinking | 阿西人工智能 | Text | - | 524.3K | 262.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | 10.8 | OfficialNo priceLowestNo price | 2026-04-01 | |
Reasoning-optimized 398B MoE agent model with extended thinking for long-horizon and multi-turn tool use Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Trinity Large Previewtrinity-large-preview | 阿西人工智能 | Text | - | 524.3K | 262.1K | Text: Supported | Text: Supported | Reason: Not supportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowestNo price | 2026-01-27 | |
Lightly post-trained 398B MoE chat model for creative work, long-context prompts, and tool-using agents Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Trinity Minitrinity-mini | 阿西人工智能 | Text | - | 131.1K | 131.1K | Text: Supported | Text: Supported | Reason: SupportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowestNo price | 2025-12-01 | |
Reasoning-tuned 26B MoE model with 3B active parameters for agents, tools, and multi-step workloads Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: SupportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Trinity Nano Previewtrinity-nano-preview | 阿西人工智能 | Text | - | 131.1K | 131.1K | Text: Supported | Text: Supported | Reason: Not supportedTools: SupportedJSON: UnknownTemperature: SupportedOpen weights: Supported | - | OfficialNo priceLowestNo price | 2025-12-01 | |
Experimental chat-tuned 6B MoE model with 1B active parameters for low-resource chat and instruction following Basics
Limits
InputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported OutputText: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported API capabilitiesReasoning: Not supportedTool call: SupportedStructured output: UnknownTemperature: Supported Pricing
| ||||||||||||
Capabilities and limits come from public sources. “Unknown” means no reliable data yet, not “not supported”. Relay counts include only sites with a current public price, de-duplicated per site.
Data source: models.dev · OpenRouter
Intelligence scores come from the OpenRouter benchmark rankings (Intelligence Index); higher is smarter.