Enter opens the highlighted result. Esc closes.

Llama 3.3 70B Instruct vs Qwen2.5 72B Instruct

Llama 3.3 70B Instruct and Qwen2.5 72B Instruct tie on the Artificial Analysis Intelligence Index at 7.7, and Qwen2.5 72B Instruct costs 1.8x less per output token ($0.40 vs $0.72 per million). Updated October 22, 2026; 46 benchmarks have a result for both.

Replace Llama 3.3 70B Instruct Replace Qwen2.5 72B Instruct
Llama 3.3 70B InstructMeta Llama
Coding
No data
Overall
26.0
Output $/M
$0.72
17
Qwen2.5 72B InstructQwen
Coding
No data
Overall
26.5
Output $/M
$0.40
27
vs boards won of 46 shared, 2 tied

Llama 3.3 70B Instruct or Qwen2.5 72B Instruct: which to pick

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

Pick Llama 3.3 70B Instruct for a larger context window (131K against 33K tokens) and work measured by Terminal-Bench Hard (Artificial Analysis run) or WeirdML, where it leads.

Model ALlama 3.3 70B InstructMeta LlamaNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen2.5 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
46
Llama 3.3 70B Instruct only
6
Qwen2.5 72B Instruct only
7
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
Llama 3.3 70B Instruct not benchmarked, Qwen2.5 72B Instruct not benchmarked.
Widest category
Agentic work, 7.3 points apart on average over 2 shared benchmarks.
Widest single gap
MATH level 5: Llama 3.3 70B Instruct 41.6% against Qwen2.5 72B Instruct 63.2%, 21.6 points apart on a 0 to 100 scale.
Output price
Llama 3.3 70B Instruct $0.72 against Qwen2.5 72B Instruct $0.40 per million output tokens, 1.8x apart.
Context window
Llama 3.3 70B Instruct takes 131K 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
CategoryLlama 3.3 70B InstructBarsQwen2.5 72B InstructTies
Coding14
Intelligence011 tie
Agentic work11
Reasoning141 tie
Math06
Knowledge33
General118
All shared boards17272 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index14
Coding benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
AA Coding Index11.9Not measuredOnly Llama 3.3 70B Instruct measured
LiveCodeBench (Artificial Analysis run)28.8%27.6%Llama 3.3 70B Instruct
Terminal-Bench Hard (Artificial Analysis run)3.0%4.5%Qwen2.5 72B Instruct
WeirdML14.4%16.0%Qwen2.5 72B Instruct
LMArena Text: coding12681292Qwen2.5 72B Instruct
LMArena Text: software and IT12731292Qwen2.5 72B Instruct
Neither model has a result on the other 33 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
AA Intelligence Index7.77.7Tie
Epoch Capabilities Index127.3129.0Qwen2.5 72B Instruct
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
τ²-Bench Telecom26.6%34.5%Qwen2.5 72B Instruct
METR time horizonsNot measured35.8%Only Qwen2.5 72B Instruct measured
BALROG23.0%16.2%Llama 3.3 70B Instruct
TheAgentCompanyNot measured5.7%Only Qwen2.5 72B Instruct measured
Berkeley Function Calling Leaderboard (BFCL)31.9%Not measuredOnly Llama 3.3 70B Instruct measured
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index14
Reasoning benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
GPQA Diamond (Artificial Analysis run)49.8%49.1%Llama 3.3 70B Instruct
Humanity's Last Exam (Artificial Analysis run)3.6%3.6%Tie
GPQA Diamond (Epoch AI run)47.4%49.1%Qwen2.5 72B Instruct
SimpleBench19.9%Not measuredOnly Llama 3.3 70B Instruct measured
BIG-Bench HardNot measured79.8%Only Qwen2.5 72B Instruct measured
LMArena Text: hard prompts12571271Qwen2.5 72B Instruct
LMArena Text: expert12251245Qwen2.5 72B Instruct
LMArena Text: hard prompts in English12771279Qwen2.5 72B Instruct
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
AA Math Index7.714.0Qwen2.5 72B Instruct
AIME 2025 (Artificial Analysis run)7.7%14.0%Qwen2.5 72B Instruct
OTIS Mock AIME5.1%8.1%Qwen2.5 72B Instruct
MATH level 541.6%63.2%Qwen2.5 72B Instruct
LMArena Text: math12671283Qwen2.5 72B Instruct
LMArena Text: mathematical work12671290Qwen2.5 72B Instruct
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index33
Knowledge benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
MMLU Pro (Artificial Analysis run)71.3%72.0%Qwen2.5 72B Instruct
MMLU86.3%85.3%Llama 3.3 70B Instruct
AI2 Reasoning ChallengeNot measured94.5%Only Qwen2.5 72B Instruct measured
LMArena Text: science12841275Llama 3.3 70B Instruct
LMArena Text: medicine and healthcare12751258Llama 3.3 70B Instruct
LMArena Text: legal and government12841288Qwen2.5 72B Instruct
LMArena Text: business and finance12431253Qwen2.5 72B Instruct
Neither model has a result on the other 11 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index118
General benchmarks for Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 72B InstructLeader
IFBench47.1%36.9%Llama 3.3 70B Instruct
Long context reasoning (AA-LCR)15.7%Not measuredOnly Llama 3.3 70B Instruct measured
DTBench59.5%62.9%Qwen2.5 72B Instruct
LMCA17.5%13.4%Llama 3.3 70B Instruct
Fiction.LiveBench33.3%Not measuredOnly Llama 3.3 70B Instruct 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 Text12741269Llama 3.3 70B Instruct
LMArena Text: instruction following12421254Qwen2.5 72B Instruct
LMArena Text: creative writing12501221Llama 3.3 70B Instruct
LMArena Text: English13031283Llama 3.3 70B Instruct
LMArena Text: non-English12361252Qwen2.5 72B Instruct
LMArena Text: Chinese12171273Qwen2.5 72B Instruct
LMArena Text: German12511234Llama 3.3 70B Instruct
LMArena Text: Japanese11501180Qwen2.5 72B Instruct
LMArena Text: Korean11421188Qwen2.5 72B Instruct
LMArena Text: French12821280Llama 3.3 70B Instruct
LMArena Text: Spanish12691255Llama 3.3 70B Instruct
LMArena Text: Russian12521264Qwen2.5 72B Instruct
LMArena Text: Polish1259Not measuredOnly Llama 3.3 70B Instruct measured
LMArena Text: multi-turn12801272Llama 3.3 70B Instruct
LMArena Text: longer queries12551281Qwen2.5 72B Instruct
LMArena Text: writing and language12551247Llama 3.3 70B Instruct
LMArena Text: entertainment and media12411211Llama 3.3 70B Instruct
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 Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
BenchmarkLlama 3.3 70B InstructQwen2.5 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

Llama 3.3 70B Instruct$0.0008Fits 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

Llama 3.3 70B Instruct$0.025Fits in one request
Qwen2.5 72B Instruct$0.019Exceeds the context window

Qwen2.5 72B Instruct costs less

Agent loop

200K input + 10K output tokens

Llama 3.3 70B Instruct$0.097Exceeds the context window
Qwen2.5 72B Instruct$0.076Exceeds the context window

Qwen2.5 72B Instruct 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 Llama 3.3 70B Instruct and Qwen2.5 72B Instruct
SpecificationMeta Llama 3.3 70B InstructQwen2.5 72B Instruct
MakerMeta LlamaQwen
WeightsOpen weightsOpen weights
ReleasedDecember 6, 2024September 19, 2024
Context window131K33K
Input $/Mtok$0.45$0.36
Output $/Mtok$0.72$0.40
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Llama 3.3 70B Instruct better than Qwen2.5 72B Instruct for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Llama 3.3 70B Instruct for a larger context window (131K against 33K tokens) and work measured by Terminal-Bench Hard (Artificial Analysis run) or WeirdML, where it leads.

Which is cheaper, Llama 3.3 70B Instruct or Qwen2.5 72B Instruct?

Llama 3.3 70B Instruct costs $0.45 per million input tokens and $0.72 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, Llama 3.3 70B Instruct or Qwen2.5 72B Instruct?

Llama 3.3 70B Instruct, with 131K tokens against 33K.

How many benchmarks is the Llama 3.3 70B Instruct vs Qwen2.5 72B Instruct comparison based on?

46 benchmarks have a result for both models as of October 22, 2026. Llama 3.3 70B Instruct's coding index rests on 0 independent benchmarks 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