Enter opens the highlighted result. Esc closes.

R1 0528 vs Gemma 3 12B

R1 0528 leads Gemma 3 12B on the Artificial Analysis Intelligence Index, 13.1 to 3.8, and Gemma 3 12B costs 13.4x less per output token ($0.1625 vs $2.18 per million). Updated October 22, 2026; 35 benchmarks have a result for both.

Replace R1 0528 Replace Gemma 3 12B
R1 0528DeepSeek
Coding
No data
Overall
46.8
Output $/M
$2.18
35
Gemma 3 12BGoogle
Coding
No data
Overall
28.0
Output $/M
$0.1625
0
vs boards won of 35 shared

R1 0528 or Gemma 3 12B: which to pick

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

Pick R1 0528 for a larger context window (164K against 131K tokens).

Model AR1 0528DeepSeekNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGemma 3 12BGoogleNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
35
R1 0528 only
25
Gemma 3 12B only
8
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
R1 0528 not benchmarked, Gemma 3 12B not benchmarked.
Widest category
Math, 38.7 points apart on average over 5 shared benchmarks.
Widest single gap
LiveCodeBench (Artificial Analysis run): R1 0528 77.0% against Gemma 3 12B 13.7%, 63.3 points apart on a 0 to 100 scale.
Output price
R1 0528 $2.18 against Gemma 3 12B $0.1625 per million output tokens, 13.4x apart.
Context window
R1 0528 takes 164K tokens, Gemma 3 12B 131K.

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
CategoryR1 0528BarsGemma 3 12BTies
Coding40
Intelligence10
Agentic work10
Reasoning60
Math50
Knowledge50
General130
All shared boards350
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
Coding benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
Aider polyglot71.4%Not measuredOnly R1 0528 measured
AA Coding IndexNot measured5.8Only Gemma 3 12B measured
LiveCodeBench (Artificial Analysis run)77.0%13.7%R1 0528
SciCodeNot measured16.4%Only Gemma 3 12B measured
Terminal-Bench Hard (Artificial Analysis run)15.9%0.8%R1 0528
WeirdML41.6%Not measuredOnly R1 0528 measured
ALE-Bench804.1Not measuredOnly R1 0528 measured
LMArena Text: coding14271281R1 0528
LMArena Text: software and IT14351313R1 0528
SciCodeNot measured0.0%Only Gemma 3 12B measured
Neither model has a result on the other 29 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
AA Intelligence Index13.13.8R1 0528
Epoch Capabilities IndexNot measured123.5Only Gemma 3 12B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
AA Agentic IndexNot measured0.1Only Gemma 3 12B measured
τ²-Bench Telecom36.5%10.8%R1 0528
METR time horizons53.8%Not measuredOnly R1 0528 measured
DeepResearch Bench35.1%Not measuredOnly R1 0528 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured30.4%Only Gemma 3 12B measured
BrowseComp-Plus16.4%Not measuredOnly R1 0528 measured
Neither model has a result on the other 35 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
GPQA Diamond (Artificial Analysis run)81.3%34.9%R1 0528
Humanity's Last Exam (Artificial Analysis run)15.8%4.2%R1 0528
GPQA Diamond (Epoch AI run)76.3%39.5%R1 0528
ARC-AGI21.2%Not measuredOnly R1 0528 measured
ARC-AGI-21.1%Not measuredOnly R1 0528 measured
SimpleBench40.8%Not measuredOnly R1 0528 measured
LMArena Text: hard prompts14151309R1 0528
LMArena Text: expert13941248R1 0528
LMArena Text: hard prompts in English14221318R1 0528
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index50
Math benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
AA Math Index76.018.3R1 0528
AIME 2025 (Artificial Analysis run)76.0%18.3%R1 0528
OTIS Mock AIME66.4%16.7%R1 0528
MATH level 596.6%Not measuredOnly R1 0528 measured
LMArena Text: math14001307R1 0528
LMArena Text: mathematical work14151336R1 0528
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
Knowledge benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
MMLU Pro (Artificial Analysis run)84.9%59.5%R1 0528
LMArena Text: science14301366R1 0528
LMArena Text: medicine and healthcare14451334R1 0528
LMArena Text: legal and government14191381R1 0528
LMArena Text: business and finance14021376R1 0528
Neither model has a result on the other 13 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Web design benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
Design Arena: websites1149Not measuredOnly R1 0528 measured
Design Arena: UI components1100Not measuredOnly R1 0528 measured
Design Arena: game dev1097Not measuredOnly R1 0528 measured
Design Arena: data viz1182Not measuredOnly R1 0528 measured
Design Arena: 3D1110Not measuredOnly R1 0528 measured
Design Arena: all code categories1138Not measuredOnly R1 0528 measured
Design Arena: SVG1038Not measuredOnly R1 0528 measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index130
General benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
IFBench39.6%36.7%R1 0528
Long context reasoning (AA-LCR)55.7%8.3%R1 0528
DTBenchNot measured48.8%Only Gemma 3 12B measured
LMCANot measured4.5%Only Gemma 3 12B measured
Fiction.LiveBench75.0%Not measuredOnly R1 0528 measured
Creative writing (Lech Mazur)81.9%Not measuredOnly R1 0528 measured
LMArena Text14281334R1 0528
LMArena Text: instruction following13821299R1 0528
LMArena Text: creative writing14051331R1 0528
LMArena Text: English14371349R1 0528
LMArena Text: non-English14121318R1 0528
LMArena Text: Chinese1442Not measuredOnly R1 0528 measured
LMArena Text: German14041370R1 0528
LMArena Text: Japanese1391Not measuredOnly R1 0528 measured
LMArena Text: Korean1360Not measuredOnly R1 0528 measured
LMArena Text: French1417Not measuredOnly R1 0528 measured
LMArena Text: Spanish1411Not measuredOnly R1 0528 measured
LMArena Text: Russian14231335R1 0528
LMArena Text: Polish1428Not measuredOnly R1 0528 measured
LMArena Text: multi-turn14051334R1 0528
LMArena Text: longer queries13911317R1 0528
LMArena Text: writing and language14021307R1 0528
LMArena Text: entertainment and media14001297R1 0528
Neither model has a result on the other 23 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for R1 0528 and Gemma 3 12B
BenchmarkR1 0528Gemma 3 12BLeader
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

R1 0528$0.0016Fits in one request
Gemma 3 12B$0.0002Fits in one request

Gemma 3 12B costs less

Repository review

50K input + 3K output tokens

R1 0528$0.032Fits in one request
Gemma 3 12B$0.0042Fits in one request

Gemma 3 12B costs less

Agent loop

200K input + 10K output tokens

R1 0528$0.122Exceeds the context window
Gemma 3 12B$0.017Exceeds the context window

Gemma 3 12B 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 R1 0528 and Gemma 3 12B
SpecificationDeepSeek R1 0528Google Gemma 3 12B
MakerDeepSeekGoogle
WeightsOpen weightsOpen weights
ReleasedMay 28, 2025March 13, 2025
Context window164K131K
Input $/Mtok$0.50$0.075
Output $/Mtok$2.18$0.1625
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is R1 0528 better than Gemma 3 12B for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick R1 0528 for a larger context window (164K against 131K tokens).

Which is cheaper, R1 0528 or Gemma 3 12B?

R1 0528 costs $0.50 per million input tokens and $2.18 per million output tokens; Gemma 3 12B costs $0.075 and $0.1625. Cost per solved task is known for neither of them.

Which has the bigger context window, R1 0528 or Gemma 3 12B?

R1 0528, with 164K tokens against 131K.

How many benchmarks is the R1 0528 vs Gemma 3 12B comparison based on?

35 benchmarks have a result for both models as of October 22, 2026. R1 0528's coding index rests on 0 independent benchmarks and Gemma 3 12B's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next