1 / 3
Swipe to compare

Trinity Large Thinking is an open-source reasoning model from Arcee AI, built on a 398B-parameter sparse Mixture-of-Experts architecture that activates approximately 13B parameters per token. Post-trained with extended chain-of-thought reasoning and agentic reinforcement learning, it achieves state-of-the-art results on agentic benchmarks including τ²-Bench (94.7%) and PinchBench (91.9%). Released under the Apache 2.0 license, it offers frontier-level tool use and multi-turn conversation capabilities that can be run fully locally or via hosted API.

Author
Arcee AIArcee AI
Release Date
2026-04-01
Knowledge Cutoff
2024
License
Open Model
I/O Format
Context Length
262K / 262K
API I/O (1M)
$0.22 / $0.85
How to Use
Output Speed
129 tok/s
Arena Overall
1375
Intelligence Index
31.9
Coding Index
27.2
Math Index
LiveBench
30.4
ForecastBench
GPQA Diamond
75.2%
HLE
14.7%
MMLU-Pro
AIME 2025
MATH-500
LB Reasoning
20.6
LB Math
44.9
LB Data Analysis
40.3
LiveCodeBench
LB Coding
65.7
LB Agentic
3.3
TAU2
90.1%
TerminalBench
22.7%
SciCode
36.1%
IFBench
56.3%
AA-LCR
0.3
Hallucination (HHEM)
6.9%
Factual Consistency (HHEM)
93.1%
LB Language
42.1
LB Instruction Following
12.2