Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

Arcee AI: trinity-large-thinking

14 benchmarks
CompanyArcee AI
Country of originUSA

Leaderboard position

#361
Overall rank (of 1126 models)
1384
Elo rating
14
Benchmarks
Top 32.1%
Percentile
In Arcee AI (of 3 models)

Benchmark highlights

Best benchmark
YC-Bench 32667

Nearest competitor from another company

The model from another company with the most similar benchmark scores.

appeared as nearest in 2 of 14 benchmarks

2/14 14.3%

Benchmarks

Predicted scores

Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.

Benchmark Predicted Confidence Based on
HMMT Feb 2026 ~75.84 64% Qwen: Qwen3.5 9B, NVIDIA Corporation: nvidia-nemotron-3-super-120b-a12b, StepFun: step-3.5-flash
LiveCodeBench v6 ~69.72 64% Qwen: Qwen3.5 9B, Cohere Labs: north-mini-code-1.0, LG AI Research: k-exaone-236b-a23b
SuperGPQA ~55.03 64% OpenAI: gpt-oss-120b, Qwen: Qwen3.5 9B, Qwen: qwen3.5-122b-a10b
Humanity's Last Exam (no tools) ~20.19 63% OpenAI: gpt-oss-120b, SpaceXAI: Grok 3 Mini, LLM360: k2-think-v2
AIME 2026 ~85.26 63% Qwen: Qwen3.5 9B, StepFun: step-3.5-flash, xhtoken: spark-x2.6-4b
IFEval ~90.24 63% Qwen: Qwen3.5 9B, orcarouter: orcacyber-nano-1.0, xhtoken: spark-x2.6-4b
SWE-bench Pro ~33.67 63% OpenAI: gpt-oss-120b, Cohere Labs: north-mini-code-1.0, xhtoken: spark-x2.6-4b
SciCode ~35.1 63% SpaceXAI: Grok 3 Mini, Cohere Labs: north-mini-code-1.0, LLM360: k2-think-v2
Humanity's Last Exam (with tools) ~28.99 63% OpenAI: gpt-oss-120b, Moonshot AI: Kimi K2 Thinking, Google: Gemma 4 31B Instruct
Artificial Analysis Intelligence Index v4.3.2 ~20.52 63% Qwen: Qwen3.5 9B, Xiaomi: mimo-v2-flash, Google: Gemma 4 31B Instruct
CowQA ~86.58 63% Qwen: Qwen3.5 9B, Qwen: Qwen3.5 35B A3B, Xiaomi: mimo-v2-flash
ScreenSpot-Pro ~65.28 63% Qwen: Qwen3.5 9B, Qwen: Qwen3.5 4B, Qwen: Qwen3.5 35B A3B
Terminal-Bench v2.1 ~41.51 62% orcarouter: orcacyber-nano-1.0, inclusionAI: ling-3.0-tiny, diffuspace: clover-35b-a3b-preview
AA-LCR ~54.98 62% SpaceXAI: Grok 3 Mini, inclusionAI: ling-3.0-tiny, diffuspace: clover-35b-a3b-preview
MMMU-Pro ~65.71 62% Google: Gemma 4 31B Instruct, Intern Language Models: intern-s1-pro, Qwen: qwen3.6-35b-a3b
GPQA ~68.43 62% Moonshot AI: Kimi K2 Thinking, xhtoken: spark-x2.6-4b, DeepSeek: deepseek-v3.2
ParseBench ~51.81 62% Google: Gemma 4 26B A4B Instruct, Qwen: Qwen3.5 9B, Qwen: Qwen3.5 4B
Terminal-Bench v2.0 ~34.74 62% NVIDIA Corporation: nvidia-nemotron-3-super-120b-a12b, Cohere Labs: north-mini-code-1.0, StepFun: step-3.5-flash
MATH-500 ~96.28 61% orcarouter: orcacyber-nano-1.0, diffuspace: clover-35b-a3b-preview, OpenBMB: minicpm5-2b
MMLU ~77.96 61% Qwen: Qwen3.5 9B, Qwen: Qwen3.5 4B, DeepSeek: DeepSeek R1
-(CoderPPL) ~-1.69 61% Qwen: Qwen3.5 9B, Qwen: Qwen3.5 4B, OpenAI: gpt-oss-20b
Claw-Eval ~51.2 61% NVIDIA Corporation: nvidia-nemotron-3-super-120b-a12b, DeepSeek: deepseek-v3.2, Qwen: qwen3.5-397b-a17b
Artificial Analysis Intelligence Index v4.1.1 (Maximum Reasoning) ~26.76 61% OpenAI: gpt-oss-120b, Google: Gemma 4 31B Instruct, Qwen: qwen3-max
Artificial Analysis Agentic Index (Maximum Reasoning) ~33.16 61% Google: Gemma 4 26B A4B Instruct, SpaceXAI: Grok 4.1 Fast, DeepSeek: DeepSeek R1
LLMArena (llmarena.ru) (Snapshot from 19 june 2026) ~1051.36 61% OpenAI: gpt-oss-120b, Xiaomi: mimo-v2-flash, DeepSeek: deepseek-v3.2