Enter opens the highlighted result. Esc closes.

Trinity Large Thinking vs Qwen2.5 72B Instruct

Trinity Large Thinking leads Qwen2.5 72B Instruct on the Artificial Analysis Intelligence Index, 10.8 to 7.7, and Qwen2.5 72B Instruct costs 2.0x less per output token ($0.40 vs $0.80 per million). Updated October 21, 2026; 33 benchmarks have a result for both.

Replace Trinity Large Thinking Replace Qwen2.5 72B Instruct
Trinity Large ThinkingArcee AI
Coding
No data
Overall
10.8
Output $/M
$0.80
33
Qwen2.5 72B InstructQwen
Coding
No data
Overall
7.7
Output $/M
$0.40
0
vs boards won of 33 shared

Trinity Large Thinking or Qwen2.5 72B Instruct: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick Trinity Large Thinking for a larger context window (262K against 33K tokens).

Model ATrinity Large ThinkingArcee AINot benchmarked1 independent benchmarkNo cost per solved taskModel BQwen2.5 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
33
Trinity Large Thinking only
19
Qwen2.5 72B Instruct only
20
Like-for-like categories
7 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Trinity Large Thinking not benchmarked, Qwen2.5 72B Instruct not benchmarked.
Widest category
Agentic work, 55.6 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: Trinity Large Thinking 90.1% against Qwen2.5 72B Instruct 34.5%, 55.6 points apart on a 0 to 100 scale.
Output price
Trinity Large Thinking $0.80 against Qwen2.5 72B Instruct $0.40 per million output tokens, 2.0x apart.
Context window
Trinity Large Thinking takes 262K tokens, Qwen2.5 72B Instruct 33K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryTrinity Large ThinkingBarsQwen2.5 72B InstructTies
Coding30
Intelligence10
Agentic work10
Reasoning50
Math20
Knowledge40
General170
All shared boards330
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
Coding benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
LMArena WebDev1238Not measuredOnly Trinity Large Thinking measured
AA Coding Index25.8Not measuredOnly Trinity Large Thinking measured
LiveCodeBench (Artificial Analysis run)Not measured27.6%Only Qwen2.5 72B Instruct measured
SciCode40.6%Not measuredOnly Trinity Large Thinking measured
Terminal-Bench Hard (Artificial Analysis run)22.7%4.5%Trinity Large Thinking
WeirdMLNot measured16.0%Only Qwen2.5 72B Instruct measured
LMArena Text: coding13571292Trinity Large Thinking
LMArena Text: software and IT13621292Trinity Large Thinking
LMArena WebDev: React1231Not measuredOnly Trinity Large Thinking measured
LMArena WebDev: HTML1240Not measuredOnly Trinity Large Thinking measured
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
AA Intelligence Index10.87.7Trinity Large Thinking
Epoch Capabilities IndexNot measured129.0Only Qwen2.5 72B Instruct measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
AA Agentic Index1.0Not measuredOnly Trinity Large Thinking measured
τ²-Bench Telecom90.1%34.5%Trinity Large Thinking
METR time horizonsNot measured35.8%Only Qwen2.5 72B Instruct measured
BALROGNot measured16.2%Only Qwen2.5 72B Instruct measured
TheAgentCompanyNot measured5.7%Only Qwen2.5 72B Instruct measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
GPQA Diamond (Artificial Analysis run)75.2%49.1%Trinity Large Thinking
Humanity's Last Exam (Artificial Analysis run)15.8%3.6%Trinity Large Thinking
GPQA Diamond (Epoch AI run)Not measured49.1%Only Qwen2.5 72B Instruct measured
BIG-Bench HardNot measured79.8%Only Qwen2.5 72B Instruct measured
LMArena Text: hard prompts13421271Trinity Large Thinking
LMArena Text: expert13591245Trinity Large Thinking
LMArena Text: hard prompts in English13561279Trinity Large Thinking
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
AA Math IndexNot measured14.0Only Qwen2.5 72B Instruct measured
AIME 2025 (Artificial Analysis run)Not measured14.0%Only Qwen2.5 72B Instruct measured
OTIS Mock AIMENot measured8.1%Only Qwen2.5 72B Instruct measured
MATH level 5Not measured63.2%Only Qwen2.5 72B Instruct measured
LMArena Text: math13651283Trinity Large Thinking
LMArena Text: mathematical work13501290Trinity Large Thinking
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
CaseLaw (Vals AI)57.9%Not measuredOnly Trinity Large Thinking measured
CorpFin (Vals AI)54.7%Not measuredOnly Trinity Large Thinking measured
MMLU Pro (Artificial Analysis run)Not measured72.0%Only Qwen2.5 72B Instruct measured
MMLUNot measured85.3%Only Qwen2.5 72B Instruct measured
AI2 Reasoning ChallengeNot measured94.5%Only Qwen2.5 72B Instruct measured
LMArena Text: science13551275Trinity Large Thinking
LMArena Text: medicine and healthcare13491258Trinity Large Thinking
LMArena Text: legal and government13451288Trinity Large Thinking
LMArena Text: business and finance13291253Trinity Large Thinking
Neither model has a result on the other 8 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
Design Arena: websites1137Not measuredOnly Trinity Large Thinking measured
Design Arena: UI components1045Not measuredOnly Trinity Large Thinking measured
Design Arena: game dev1079Not measuredOnly Trinity Large Thinking measured
Design Arena: data viz1098Not measuredOnly Trinity Large Thinking measured
Design Arena: 3D1082Not measuredOnly Trinity Large Thinking measured
Design Arena: all code categories1117Not measuredOnly Trinity Large Thinking measured
Design Arena: SVG1017Not measuredOnly Trinity Large Thinking measured
Design Arena: ASCII art1041Not measuredOnly Trinity Large Thinking measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index170
General benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
IFBench56.3%36.9%Trinity Large Thinking
Long context reasoning (AA-LCR)38.0%Not measuredOnly Trinity Large Thinking measured
DTBenchNot measured62.9%Only Qwen2.5 72B Instruct measured
LMCANot measured13.4%Only Qwen2.5 72B Instruct measured
Surface Evolver Bench15.6%Not measuredOnly Trinity Large Thinking measured
HellaSwagNot measured84.8%Only Qwen2.5 72B Instruct measured
WinoGrandeNot measured82.3%Only Qwen2.5 72B Instruct measured
PIQANot measured82.6%Only Qwen2.5 72B Instruct measured
LMArena Text13401269Trinity Large Thinking
LMArena Text: instruction following13261254Trinity Large Thinking
LMArena Text: creative writing13081221Trinity Large Thinking
LMArena Text: English13561283Trinity Large Thinking
LMArena Text: non-English13251252Trinity Large Thinking
LMArena Text: Chinese13711273Trinity Large Thinking
LMArena Text: German13561234Trinity Large Thinking
LMArena Text: Japanese13121180Trinity Large Thinking
LMArena Text: Korean13061188Trinity Large Thinking
LMArena Text: French13661280Trinity Large Thinking
LMArena Text: Spanish13271255Trinity Large Thinking
LMArena Text: Russian13381264Trinity Large Thinking
LMArena Text: Polish1312Not measuredOnly Trinity Large Thinking measured
LMArena Text: multi-turn13201272Trinity Large Thinking
LMArena Text: longer queries13181281Trinity Large Thinking
LMArena Text: writing and language13171247Trinity Large Thinking
LMArena Text: entertainment and media12961211Trinity Large Thinking
Neither model has a result on the other 21 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Trinity Large Thinking and Qwen2.5 72B Instruct
BenchmarkTrinity Large ThinkingQwen2.5 72B InstructLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Trinity Large Thinking$0.0006Fits in one request
Qwen2.5 72B Instruct$0.0006Fits in one request

Qwen2.5 72B Instruct costs less

Repository review

50K input + 3K output tokens

Trinity Large Thinking$0.015Fits in one request
Qwen2.5 72B Instruct$0.019Exceeds the context window

Trinity Large Thinking costs less

Agent loop

200K input + 10K output tokens

Trinity Large Thinking$0.058Fits in one request
Qwen2.5 72B Instruct$0.076Exceeds the context window

Trinity Large Thinking costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Trinity Large Thinking and Qwen2.5 72B Instruct
SpecificationArcee AI Trinity Large ThinkingQwen2.5 72B Instruct
MakerArcee AIQwen
WeightsOpen weightsOpen weights
ReleasedApril 1, 2026September 19, 2024
Context window262K33K
Input $/Mtok$0.25$0.36
Output $/Mtok$0.80$0.40
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it10
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Trinity Large Thinking better than Qwen2.5 72B Instruct for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Trinity Large Thinking for a larger context window (262K against 33K tokens).

Which is cheaper, Trinity Large Thinking or Qwen2.5 72B Instruct?

Trinity Large Thinking costs $0.25 per million input tokens and $0.80 per million output tokens; Qwen2.5 72B Instruct costs $0.36 and $0.40. Cost per solved task is known for neither of them.

Which has the bigger context window, Trinity Large Thinking or Qwen2.5 72B Instruct?

Trinity Large Thinking, with 262K tokens against 33K.

How many benchmarks is the Trinity Large Thinking vs Qwen2.5 72B Instruct comparison based on?

33 benchmarks have a result for both models as of October 21, 2026. Trinity Large Thinking's coding index rests on 1 independent benchmark and Qwen2.5 72B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next