GeekRadar / The comparison desk

COMPARE.

AI models

Not reported

Trinity Large Thinking

Model evaluation

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Trinity Large Thinking 1340.3 Trinity Large Thinking · 95% interval 1335.6–1344.9 · 30,290 votes · 2026-10-28

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,371.43 rating

Confidence interval: 1,357.40 – 1,385.46

2078 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,357.12 rating

Confidence interval: 1,349.68 – 1,364.55

8538 votes

Oct 28, 2026

Category: coding
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,308.33 rating

Confidence interval: 1,298.99 – 1,317.68

4725 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,355.89 rating

Confidence interval: 1,349.84 – 1,361.95

14359 votes

Oct 28, 2026

Category: english
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,303.34 rating

Confidence interval: 1,297.19 – 1,309.49

23507 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,359.14 rating

Confidence interval: 1,347.49 – 1,370.79

2907 votes

Oct 28, 2026

Category: expert
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,366.79 rating

Confidence interval: 1,346.46 – 1,387.11

1031 votes

Oct 28, 2026

Category: french
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,355.55 rating

Confidence interval: 1,326.80 – 1,384.31

408 votes

Oct 28, 2026

Category: german
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,341.94 rating

Confidence interval: 1,336.36 – 1,347.51

20031 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,356.18 rating

Confidence interval: 1,349.17 – 1,363.18

9940 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,329.58 rating

Confidence interval: 1,321.18 – 1,337.97

6114 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,296.30 rating

Confidence interval: 1,287.86 – 1,304.73

6244 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,345.80 rating

Confidence interval: 1,332.95 – 1,358.64

2387 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,355.35 rating

Confidence interval: 1,346.35 – 1,364.34

5099 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,351.04 rating

Confidence interval: 1,335.68 – 1,366.40

1750 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,349.83 rating

Confidence interval: 1,336.59 – 1,363.08

2280 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,362.04 rating

Confidence interval: 1,355.60 – 1,368.48

12275 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,317.64 rating

Confidence interval: 1,309.88 – 1,325.40

7265 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,326.16 rating

Confidence interval: 1,319.33 – 1,332.99

10332 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / japanesehuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,312.06 rating

Confidence interval: 1,275.68 – 1,348.43

282 votes

Oct 28, 2026

Category: japanese
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,304.77 rating

Confidence interval: 1,278.59 – 1,330.95

534 votes

Oct 28, 2026

Category: korean
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,318.17 rating

Confidence interval: 1,311.50 – 1,324.84

13067 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,365.45 rating

Confidence interval: 1,350.62 – 1,380.29

1720 votes

Oct 28, 2026

Category: math
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,320.33 rating

Confidence interval: 1,311.42 – 1,329.24

5597 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,325.32 rating

Confidence interval: 1,319.55 – 1,331.09

15931 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,340.28 rating

Confidence interval: 1,335.65 – 1,344.92

30290 votes

Oct 28, 2026

Category: overall
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,311.51 rating

Confidence interval: 1,285.13 – 1,337.89

566 votes

Oct 28, 2026

Category: polish
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,338.42 rating

Confidence interval: 1,327.43 – 1,349.41

3282 votes

Oct 28, 2026

Category: russian
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-28 · rating

Trinity Large Thinking

1,326.92 rating

Confidence interval: 1,306.32 – 1,347.52

946 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: Trinity Large Thinking
Evaluation type: human preference

Run it locally

Inference · memory estimate

Trinity Large Thinking

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Apache 2.0