Enter opens the highlighted result. Esc closes.

Gemma 4 26B A4B vs Mercury 2.5

Gemma 4 26B A4B leads Mercury 2.5 on the Artificial Analysis Intelligence Index, 16.7 to 12.3, and Mercury 2.5 costs 4.0x less per output token ($0.15 vs $0.60 per million). Updated October 22, 2026; 5 benchmarks have a result for both.

Replace Gemma 4 26B A4B Replace Mercury 2.5
Gemma 4 26B A4BGoogle
Coding
No data
Overall
54.7
Output $/M
$0.60
4
Mercury 2.5Inception
Coding
1.6
Overall
No data
Output $/M
$0.15
1
vs boards won of 5 shared

Gemma 4 26B A4B or Mercury 2.5: which to pick

Only Mercury 2.5 has a coding score, 1.6 on the Vibe Coding Index; Gemma 4 26B A4B has none, so the two cannot be ranked on coding.

Pick Gemma 4 26B A4B for a larger context window (262K against 260K tokens) and work measured by SciCode or ALE-Bench, where it leads.

Model AGemma 4 26B A4BGoogleNot benchmarked1 independent benchmarkNo cost per solved taskModel BMercury 2.5Inception1.6/1002 independent benchmarksNo cost per solved task
Shared results
5
Gemma 4 26B A4B only
50
Mercury 2.5 only
17
Like-for-like categories
4 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, Mercury 2.5 1.6 from 2 benchmarks.
Widest category
Coding, 11.2 points apart on average over 2 shared benchmarks.
Widest single gap
ALE-Bench: Gemma 4 26B A4B 927.2 against Mercury 2.5 301.6, 20.9 points apart on a 0 to 100 scale.
Output price
Gemma 4 26B A4B $0.60 against Mercury 2.5 $0.15 per million output tokens, 4.0x apart.
Context window
Gemma 4 26B A4B takes 262K tokens, Mercury 2.5 260K.

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 A4BBarsMercury 2.5Ties
Coding20
Intelligence10
Reasoning10
General01
All shared boards41
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
Coding benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
LMArena WebDev1360Not measuredOnly Gemma 4 26B A4B measured
AA Coding Index39.3Not measuredOnly Gemma 4 26B A4B measured
Terminal-Bench 4 (Vals AI)Not measured0.0%Only Mercury 2.5 measured
Vibe Code Bench (Vals AI)Not measured3.7%Only Mercury 2.5 measured
Code migration (Vals AI)Not measured4.5%Only Mercury 2.5 measured
IOI (Vals AI)Not measured2.4%Only Mercury 2.5 measured
Terminal-Bench 2.1 (Vals AI)Not measured34.5%Only Mercury 2.5 measured
SkillsBench (Vals AI)Not measured18.1%Only Mercury 2.5 measured
CyberBench (Vals AI)Not measured45.5%Only Mercury 2.5 measured
SciCode40.0%38.5%Gemma 4 26B A4B
Terminal-Bench Hard (Artificial Analysis run)25.0%Not measuredOnly Gemma 4 26B A4B measured
WeirdML35.2%Not measuredOnly Gemma 4 26B A4B measured
ALE-Bench927.2301.6Gemma 4 26B A4B
LMArena Text: coding1446Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: software and IT1454Not measuredOnly Gemma 4 26B A4B measured
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 22 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
AA Intelligence Index16.712.3Gemma 4 26B A4B
Epoch Capabilities Index141.9Not measuredOnly Gemma 4 26B A4B measured
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Agentic work benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
AA Agentic Index2.9Not measuredOnly Gemma 4 26B A4B measured
Finance Agent (Vals AI)Not measured18.9%Only Mercury 2.5 measured
Tax Agent Bench (Vals AI)Not measured2.5%Only Mercury 2.5 measured
Legal Research Bench (Vals AI)Not measured4.3%Only Mercury 2.5 measured
Excel Modeling Benchmark (Vals AI)Not measured8.8%Only Mercury 2.5 measured
Terminal-Bench-Science (Vals AI)Not measured0.0%Only Mercury 2.5 measured
τ²-Bench Telecom43.6%Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index10
Reasoning benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
GPQA Diamond (Artificial Analysis run)79.2%Not measuredOnly Gemma 4 26B A4B measured
Humanity's Last Exam (Artificial Analysis run)19.3%11.8%Gemma 4 26B A4B
GPQA Diamond (Epoch AI run)73.2%Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: hard prompts1438Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: expert1447Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: hard prompts in English1448Not measuredOnly Gemma 4 26B A4B measured
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 14 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Math benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
OTIS Mock AIME82.2%Not measuredOnly Gemma 4 26B A4B measured
ProofBenchNot measured3.0%Only Mercury 2.5 measured
LMArena Text: math1470Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: mathematical work1475Not measuredOnly Gemma 4 26B A4B measured
Neither model has a result on the other 9 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Knowledge benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
LegalBench (Vals AI)Not measured83.1%Only Mercury 2.5 measured
MedCode (Vals AI)Not measured31.3%Only Mercury 2.5 measured
LMArena Text: science1446Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: medicine and healthcare1428Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: legal and government1457Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: business and finance1429Not measuredOnly Gemma 4 26B A4B measured
FACTS Benchmark Suite42.7%Not measuredOnly Gemma 4 26B A4B measured
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 Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
General benchmarks for Gemma 4 26B A4B and Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
Public Benefits Bench (Vals AI)Not measured45.3%Only Mercury 2.5 measured
MedScribe (Vals AI)Not measured55.1%Only Mercury 2.5 measured
IFBench72.4%Not measuredOnly Gemma 4 26B A4B measured
Long context reasoning (AA-LCR)65.7%71.7%Mercury 2.5
DTBench74.9%Not measuredOnly Gemma 4 26B A4B measured
LMCA29.7%Not measuredOnly Gemma 4 26B A4B measured
LMArena Text1434Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: instruction following1420Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: creative writing1400Not measuredOnly Gemma 4 26B A4B measured
LMArena Vision1260Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: English1444Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: non-English1421Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: Chinese1495Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: French1460Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: Spanish1415Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: Russian1434Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: multi-turn1441Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: longer queries1427Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: writing and language1405Not measuredOnly Gemma 4 26B A4B measured
LMArena Text: entertainment and media1403Not measuredOnly Gemma 4 26B A4B measured
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 19 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 Mercury 2.5
BenchmarkGemma 4 26B A4BMercury 2.5Leader
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
Mercury 2.5$0.0001Fits in one request

Mercury 2.5 costs less

Repository review

50K input + 3K output tokens

Gemma 4 26B A4B$0.0093Fits in one request
Mercury 2.5$0.0024Fits in one request

Mercury 2.5 costs less

Agent loop

200K input + 10K output tokens

Gemma 4 26B A4B$0.036Fits in one request
Mercury 2.5$0.0095Fits in one request

Mercury 2.5 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 Mercury 2.5
SpecificationGoogle Gemma 4 26B A4BInception Mercury 2.5
MakerGoogleInception
WeightsOpen weightsNo public weights listed
ReleasedApril 3, 2026September 8, 2026
Context window262K260K
Input $/Mtok$0.15$0.04
Output $/Mtok$0.60$0.15
Coding indexNot benchmarked1.6
Independent measurements behind it12
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 4 26B A4B better than Mercury 2.5 for coding?

Only Mercury 2.5 has a coding score, 1.6 on the Vibe Coding Index; Gemma 4 26B A4B has none, so the two cannot be ranked on coding. Pick Gemma 4 26B A4B for a larger context window (262K against 260K tokens) and work measured by SciCode or ALE-Bench, where it leads.

Which is cheaper, Gemma 4 26B A4B or Mercury 2.5?

Gemma 4 26B A4B costs $0.15 per million input tokens and $0.60 per million output tokens; Mercury 2.5 costs $0.04 and $0.15. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemma 4 26B A4B or Mercury 2.5?

Gemma 4 26B A4B, with 262K tokens against 260K.

How many benchmarks is the Gemma 4 26B A4B vs Mercury 2.5 comparison based on?

5 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 Mercury 2.5's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next