Trinity-Large-Preview

Released
2026-01-27
Last refreshed
2026-06-29
Status
Researched 213d ago
Open sourceCommercial use: permittedRAGAgentsLong contextClassificationJSON / Tool use
Arcee AI releases · 8 in the last 12 months · this family litChangelog →
Specifications
Family
Trinity
Released
2026-01-27
Context
128K
Parameters
400B
Architecture
Mixture of Experts
Specialization
general
Openness
Open source
License
Apache 2.0OSI-approvedPermitted
Weights
Available
Code
Unknown
Training
Fine-tuned
Created by

Agentic AI workflows, efficient and secure

San Francisco, California, United States
Founded 2023
Website
Pricing

No tracked provider pricing is available yet.

About

400B sparse MoE instruct model with 13B active parameters per token, served at 128K context via 8-bit quantized API. Trained on 20T tokens. Production-ready for agentic and tool-use applications; predecessor to Trinity-Large-Thinking. Available free on OpenRouter.

Providers

No active provider routes are tracked for this model.

Deprecated provider routes(3)

ProviderStatusCatalog ID
OpenRouter
Deprecated
arcee-ai/trinity-large-preview
Arcee AI
Deprecated
—
Vercel AI Gateway
Deprecated
arcee-ai/trinity-large-preview

Capabilities

JSON / Tool useStructured Outputs

Benchmark peer barsfor RAG

No task-mapped benchmark peers are available for this model yet.

Benchmark scores(1)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
1320.0text creative_writing 2026-10-22Observed 2026-10-22
Configuration: Trinity Large Preview
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'Trinity Large Preview' (category creative_writing in text/latest-00000-of-00001.parquet): rank 200 of 416, rating 1319.6 (95% CI 1310.5-1328.7), 4767 votes. The board entry or catalog model carries a Preview/Beta/Experimental label. Source: LMArena (CC BY 4.0). Retrieved 2026-10-27.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

Migration checks

No linked migration route is available for this model yet.

Compare Trinity-Large-Preview with other models

Show all 20 popular comparisonssorted by 7-day search impressions