Enter opens the highlighted result. Esc closes.

Gemma 3 12B vs Qwen3 30B A3B Instruct 2507

Qwen3 30B A3B Instruct 2507 leads Gemma 3 12B on the Artificial Analysis Intelligence Index, 7.5 to 3.8, and Gemma 3 12B costs 1.8x less per output token ($0.1625 vs $0.30 per million). Updated October 22, 2026; 39 benchmarks have a result for both.

Replace Gemma 3 12B Replace Qwen3 30B A3B Instruct 2507
Gemma 3 12BGoogle
Coding
No data
Overall
28.0
Output $/M
$0.1625
1
Qwen3 30B A3B Instruct 2507Qwen
Coding
No data
Overall
41.5
Output $/M
$0.30
38
vs boards won of 39 shared

Gemma 3 12B or Qwen3 30B A3B Instruct 2507: which to pick

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

Pick Gemma 3 12B for output tokens 1.8x cheaper ($0.1625 vs $0.30 per million) and work measured by LiveCodeBench (Artificial Analysis run) or SciCode, where it leads.

Model AGemma 3 12BGoogleNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen3 30B A3B Instruct 2507QwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
39
Gemma 3 12B only
4
Qwen3 30B A3B Instruct 2507 only
6
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 3 12B not benchmarked, Qwen3 30B A3B Instruct 2507 not benchmarked.
Widest category
Math, 33.7 points apart on average over 5 shared benchmarks.
Widest single gap
LiveCodeBench (Artificial Analysis run): Gemma 3 12B 13.7% against Qwen3 30B A3B Instruct 2507 70.7%, 57.0 points apart on a 0 to 100 scale.
Output price
Gemma 3 12B $0.1625 against Qwen3 30B A3B Instruct 2507 $0.30 per million output tokens, 1.8x apart.
Context window
Gemma 3 12B takes 131K tokens, Qwen3 30B A3B Instruct 2507 262K.

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 3 12BBarsQwen3 30B A3B Instruct 2507Ties
Coding05
Intelligence01
Agentic work02
Reasoning06
Math05
Knowledge05
General114
All shared boards138
CodingLike-for-likeSix of these boards feed the Vibe Coding Index05
Coding benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
AA Coding Index5.8Not measuredOnly Gemma 3 12B measured
LiveCodeBench (Artificial Analysis run)13.7%70.7%Qwen3 30B A3B Instruct 2507
SciCode16.4%33.0%Qwen3 30B A3B Instruct 2507
Terminal-Bench Hard (Artificial Analysis run)0.8%6.1%Qwen3 30B A3B Instruct 2507
LMArena Text: coding12811416Qwen3 30B A3B Instruct 2507
LMArena Text: software and IT13131417Qwen3 30B A3B Instruct 2507
SciCode0.0%Not measuredOnly Gemma 3 12B measured
Neither model has a result on the other 36 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
AA Intelligence Index3.87.5Qwen3 30B A3B Instruct 2507
Epoch Capabilities Index123.5Not measuredOnly Gemma 3 12B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
AA Agentic Index0.1Not measuredOnly Gemma 3 12B measured
τ²-Bench Telecom10.8%28.1%Qwen3 30B A3B Instruct 2507
Berkeley Function Calling Leaderboard (BFCL)30.4%41.4%Qwen3 30B A3B Instruct 2507
Neither model has a result on the other 39 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
GPQA Diamond (Artificial Analysis run)34.9%70.7%Qwen3 30B A3B Instruct 2507
Humanity's Last Exam (Artificial Analysis run)4.2%10.3%Qwen3 30B A3B Instruct 2507
GPQA Diamond (Epoch AI run)39.5%55.6%Qwen3 30B A3B Instruct 2507
LMArena Text: hard prompts13091398Qwen3 30B A3B Instruct 2507
LMArena Text: expert12481397Qwen3 30B A3B Instruct 2507
LMArena Text: hard prompts in English13181406Qwen3 30B A3B Instruct 2507
Neither model has a result on the other 17 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index05
Math benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
AA Math Index18.366.3Qwen3 30B A3B Instruct 2507
AIME 2025 (Artificial Analysis run)18.3%66.3%Qwen3 30B A3B Instruct 2507
OTIS Mock AIME16.7%62.2%Qwen3 30B A3B Instruct 2507
LMArena Text: math13071394Qwen3 30B A3B Instruct 2507
LMArena Text: mathematical work13361411Qwen3 30B A3B Instruct 2507
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
MMLU Pro (Artificial Analysis run)59.5%80.5%Qwen3 30B A3B Instruct 2507
LMArena Text: science13661393Qwen3 30B A3B Instruct 2507
LMArena Text: medicine and healthcare13341392Qwen3 30B A3B Instruct 2507
LMArena Text: legal and government13811387Qwen3 30B A3B Instruct 2507
LMArena Text: business and finance13761389Qwen3 30B A3B Instruct 2507
Neither model has a result on the other 15 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index114
General benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
IFBench36.7%50.7%Qwen3 30B A3B Instruct 2507
Long context reasoning (AA-LCR)8.3%61.3%Qwen3 30B A3B Instruct 2507
DTBench48.8%67.2%Qwen3 30B A3B Instruct 2507
LMCA4.5%22.4%Qwen3 30B A3B Instruct 2507
LMArena Text13341384Qwen3 30B A3B Instruct 2507
LMArena Text: instruction following12991363Qwen3 30B A3B Instruct 2507
LMArena Text: creative writing13311317Gemma 3 12B
LMArena Text: English13491391Qwen3 30B A3B Instruct 2507
LMArena Text: non-English13181372Qwen3 30B A3B Instruct 2507
LMArena Text: ChineseNot measured1432Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: German13701379Qwen3 30B A3B Instruct 2507
LMArena Text: JapaneseNot measured1337Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: KoreanNot measured1330Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: FrenchNot measured1419Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: SpanishNot measured1404Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: Russian13351369Qwen3 30B A3B Instruct 2507
LMArena Text: PolishNot measured1353Only Qwen3 30B A3B Instruct 2507 measured
LMArena Text: multi-turn13341378Qwen3 30B A3B Instruct 2507
LMArena Text: longer queries13171378Qwen3 30B A3B Instruct 2507
LMArena Text: writing and language13071336Qwen3 30B A3B Instruct 2507
LMArena Text: entertainment and media12971334Qwen3 30B A3B Instruct 2507
Neither model has a result on the other 27 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 3 12B and Qwen3 30B A3B Instruct 2507
BenchmarkGemma 3 12BQwen3 30B A3B Instruct 2507Leader
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 3 12B$0.0002Fits in one request
Qwen3 30B A3B Instruct 2507$0.0002Fits in one request

Gemma 3 12B costs less

Repository review

50K input + 3K output tokens

Gemma 3 12B$0.0042Fits in one request
Qwen3 30B A3B Instruct 2507$0.0054Fits in one request

Gemma 3 12B costs less

Agent loop

200K input + 10K output tokens

Gemma 3 12B$0.017Exceeds the context window
Qwen3 30B A3B Instruct 2507$0.021Fits in one request

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 Gemma 3 12B and Qwen3 30B A3B Instruct 2507
SpecificationGoogle Gemma 3 12BQwen3 30B A3B Instruct 2507
MakerGoogleQwen
WeightsOpen weightsOpen weights
ReleasedMarch 13, 2025July 29, 2025
Context window131K262K
Input $/Mtok$0.075$0.09
Output $/Mtok$0.1625$0.30
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 3 12B better than Qwen3 30B A3B Instruct 2507 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Gemma 3 12B for output tokens 1.8x cheaper ($0.1625 vs $0.30 per million) and work measured by LiveCodeBench (Artificial Analysis run) or SciCode, where it leads.

Which is cheaper, Gemma 3 12B or Qwen3 30B A3B Instruct 2507?

Gemma 3 12B costs $0.075 per million input tokens and $0.1625 per million output tokens; Qwen3 30B A3B Instruct 2507 costs $0.09 and $0.30. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemma 3 12B or Qwen3 30B A3B Instruct 2507?

Qwen3 30B A3B Instruct 2507, with 262K tokens against 131K.

How many benchmarks is the Gemma 3 12B vs Qwen3 30B A3B Instruct 2507 comparison based on?

39 benchmarks have a result for both models as of October 22, 2026. Gemma 3 12B's coding index rests on 0 independent benchmarks and Qwen3 30B A3B Instruct 2507's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next