frames·cloud

Trinity-Large-Thinking

arcee-trinity-large-thinking

fields

nameREQ Trinity-Large-Thinking www.arcee.ai 2026-05-05
vendorREQ Arcee AI www.arcee.ai 2026-05-05
released_at 2026-04-01 www.arcee.ai 2026-05-05
modality text, code www.arcee.ai 2026-05-05
access open-weights www.arcee.ai 2026-05-05
context_window
parameters 399B total / 13B active www.arcee.ai 2026-05-05
benchmark_score 63.2 www.arcee.ai 2026-05-05
benchmark_name SWE-Bench Verified www.arcee.ai 2026-05-05
announcement_url https://www.arcee.ai/blog/trinity-large-thinking www.arcee.ai 2026-05-05

history · 9 fields · 9 revisions

name1 revision
Trinity-Large-Thinking current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
vendor1 revision
Arcee AI current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
released_at1 revision
2026-04-01 current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
modality1 revision
text, code current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
access1 revision
open-weights current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
parameters1 revision
399B total / 13B active current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
benchmark_score1 revision
63.2 current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
benchmark_name1 revision
SWE-Bench Verified current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.
announcement_url1 revision
https://www.arcee.ai/blog/trinity-large-thinking current www.arcee.ai · 2026-05-05
Trinity-Large-Thinking is live. A frontier open reasoning model for complex, long-horizon agents and multi-turn tool calling released under Apache 2.0. 399B total parameters, 13B active (4-of-256 experts). SWE-Bench Verified 63.2%.