Enter opens the highlighted result. Esc closes.

Gemma 4 26B A4B vs Qwen2.5 72B Instruct

Gemma 4 26B A4B leads Qwen2.5 72B Instruct on the Artificial Analysis Intelligence Index, 16.7 to 7.7, and Qwen2.5 72B Instruct costs 1.5x less per output token ($0.40 vs $0.60 per million). Updated October 22, 2026; 36 benchmarks have a result for both.

Replace Gemma 4 26B A4B Replace Qwen2.5 72B Instruct
Gemma 4 26B A4BGoogle
Coding
No data
Overall
54.7
Output $/M
$0.60
36
Qwen2.5 72B InstructQwen
Coding
No data
Overall
26.5
Output $/M
$0.40
0
vs boards won of 36 shared

Gemma 4 26B A4B or Qwen2.5 72B Instruct: which to pick

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

Pick Gemma 4 26B A4B for a larger context window (262K against 33K tokens).

Model AGemma 4 26B A4BGoogleNot benchmarked1 independent benchmarkNo cost per solved taskModel BQwen2.5 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
36
Gemma 4 26B A4B only
19
Qwen2.5 72B Instruct only
17
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
Gemma 4 26B A4B not benchmarked, Qwen2.5 72B Instruct not benchmarked.
Widest category
Math, 45.4 points apart on average over 3 shared benchmarks.
Widest single gap
OTIS Mock AIME: Gemma 4 26B A4B 82.2% against Qwen2.5 72B Instruct 8.1%, 74.2 points apart on a 0 to 100 scale.
Output price
Gemma 4 26B A4B $0.60 against Qwen2.5 72B Instruct $0.40 per million output tokens, 1.5x apart.
Context window
Gemma 4 26B A4B 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
CategoryGemma 4 26B A4BBarsQwen2.5 72B InstructTies
Coding40
Intelligence20
Agentic work10
Reasoning60
Math30
Knowledge40
General160
All shared boards360
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
Coding benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
LMArena WebDev1360Not measuredOnly Gemma 4 26B A4B measured
AA Coding Index39.3Not measuredOnly Gemma 4 26B A4B measured
LiveCodeBench (Artificial Analysis run)Not measured27.6%Only Qwen2.5 72B Instruct measured
SciCode40.0%Not measuredOnly Gemma 4 26B A4B measured
Terminal-Bench Hard (Artificial Analysis run)25.0%4.5%Gemma 4 26B A4B
WeirdML35.2%16.0%Gemma 4 26B A4B
ALE-Bench927.2Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: coding14461292Gemma 4 26B A4B
LMArena Text: software and IT14541292Gemma 4 26B A4B
LMArena WebDev: React1351Not measuredOnly Gemma 4 26B A4B measured
LMArena WebDev: HTML1362Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
AA Intelligence Index16.77.7Gemma 4 26B A4B
Epoch Capabilities Index141.9129.0Gemma 4 26B A4B
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
AA Agentic Index2.9Not measuredOnly Gemma 4 26B A4B measured
τ²-Bench Telecom43.6%34.5%Gemma 4 26B A4B
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 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
GPQA Diamond (Artificial Analysis run)79.2%49.1%Gemma 4 26B A4B
Humanity's Last Exam (Artificial Analysis run)19.3%3.6%Gemma 4 26B A4B
GPQA Diamond (Epoch AI run)73.2%49.1%Gemma 4 26B A4B
BIG-Bench HardNot measured79.8%Only Qwen2.5 72B Instruct measured
LMArena Text: hard prompts14381271Gemma 4 26B A4B
LMArena Text: expert14471245Gemma 4 26B A4B
LMArena Text: hard prompts in English14481279Gemma 4 26B A4B
LMArena Vision: diagrams1264Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: homework1293Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 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 AIME82.2%8.1%Gemma 4 26B A4B
MATH level 5Not measured63.2%Only Qwen2.5 72B Instruct measured
LMArena Text: math14701283Gemma 4 26B A4B
LMArena Text: mathematical work14751290Gemma 4 26B A4B
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 Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
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: science14461275Gemma 4 26B A4B
LMArena Text: medicine and healthcare14281258Gemma 4 26B A4B
LMArena Text: legal and government14571288Gemma 4 26B A4B
LMArena Text: business and finance14291253Gemma 4 26B A4B
FACTS Benchmark Suite42.7%Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 10 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index160
General benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 72B InstructLeader
IFBench72.4%36.9%Gemma 4 26B A4B
Long context reasoning (AA-LCR)65.7%Not measuredOnly Gemma 4 26B A4B measured
DTBench74.9%62.9%Gemma 4 26B A4B
LMCA29.7%13.4%Gemma 4 26B A4B
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 Text14341269Gemma 4 26B A4B
LMArena Text: instruction following14201254Gemma 4 26B A4B
LMArena Text: creative writing14001221Gemma 4 26B A4B
LMArena Vision1260Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: English14441283Gemma 4 26B A4B
LMArena Text: non-English14211252Gemma 4 26B A4B
LMArena Text: Chinese14951273Gemma 4 26B A4B
LMArena Text: GermanNot measured1234Only Qwen2.5 72B Instruct measured
LMArena Text: JapaneseNot measured1180Only Qwen2.5 72B Instruct measured
LMArena Text: KoreanNot measured1188Only Qwen2.5 72B Instruct measured
LMArena Text: French14601280Gemma 4 26B A4B
LMArena Text: Spanish14151255Gemma 4 26B A4B
LMArena Text: Russian14341264Gemma 4 26B A4B
LMArena Text: multi-turn14411272Gemma 4 26B A4B
LMArena Text: longer queries14271281Gemma 4 26B A4B
LMArena Text: writing and language14051247Gemma 4 26B A4B
LMArena Text: entertainment and media14031211Gemma 4 26B A4B
LMArena Vision: English1256Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: Chinese1312Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: reading text1268Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: humor1272Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: captioning1297Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: recognizing things1191Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision: creative writing1256Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 4 26B A4B and Qwen2.5 72B Instruct
BenchmarkGemma 4 26B A4BQwen2.5 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

Gemma 4 26B A4B$0.0004Fits in one request
Qwen2.5 72B Instruct$0.0006Fits in one request

Gemma 4 26B A4B costs less

Repository review

50K input + 3K output tokens

Gemma 4 26B A4B$0.0093Fits in one request
Qwen2.5 72B Instruct$0.019Exceeds the context window

Gemma 4 26B A4B costs less

Agent loop

200K input + 10K output tokens

Gemma 4 26B A4B$0.036Fits in one request
Qwen2.5 72B Instruct$0.076Exceeds the context window

Gemma 4 26B A4B 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 Gemma 4 26B A4B and Qwen2.5 72B Instruct
SpecificationGoogle Gemma 4 26B A4BQwen2.5 72B Instruct
MakerGoogleQwen
WeightsOpen weightsOpen weights
ReleasedApril 3, 2026September 19, 2024
Context window262K33K
Input $/Mtok$0.15$0.36
Output $/Mtok$0.60$0.40
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it10
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 4 26B A4B better than Qwen2.5 72B Instruct for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Gemma 4 26B A4B for a larger context window (262K against 33K tokens).

Which is cheaper, Gemma 4 26B A4B or Qwen2.5 72B Instruct?

Gemma 4 26B A4B costs $0.15 per million input tokens and $0.60 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, Gemma 4 26B A4B or Qwen2.5 72B Instruct?

Gemma 4 26B A4B, with 262K tokens against 33K.

How many benchmarks is the Gemma 4 26B A4B vs Qwen2.5 72B Instruct comparison based on?

36 benchmarks have a result for both models as of October 22, 2026. Gemma 4 26B A4B'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