Gemma 4 12B vs Nemotron 3 Nano Omni 30B A3B

Change models
IndexAll categories, out of 100.
38.3 ±3.9
38.5 ±4.5 best
GapIndex points below the top.
−0.2 · no clear gap
Top
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
—
228/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-06
2026-04
Input / 1MUS dollars per 1M input tokens.
—
—
Output / 1MUS dollars per 1M output tokens.
—
—
1K chats2K tokens in, 500 out, each.
—
—
1K documents20K tokens in, 1K out, each.
—
—
1K agent tasks200K tokens in, 20K out, each.
—
—
ContextMaximum tokens in one request.
256K
256K
AgenticMulti-step tasks with tools.
30.9
33.8 best
CodingCode writing and repair.
42.8 best
37.5
ReasoningLogic problems and puzzles.
41.2 best
35.6
MultimodalTasks with images and text.
35.4
40.6 best
KnowledgeFacts and expert knowledge.
39.5 best
38.0
MultilingualTasks in many languages.
30.0
—
InstructionTasks with strict prompt rules.
56.8 best
52.6
MathMath problems.
40.8
43.4 best

Benchmarks

11 shared
Only Gemma 4 12B: 18Only Nemotron 3 Nano Omni 30B A3B: 13
Gemma 4 12B
Nemotron 3 Nano Omni 30B A3B
15.7% best122 / 216
4.8%175 / 216
63.7% best132 / 211
39.7%162 / 211
0.0%145 / 207
0.0%145 / 207
75.3% best109 / 193
46.9%174 / 193
78.8% best71 / 144
72.2%86 / 144
36.3%102 / 143
45.3% best96 / 143
73.5% best27 / 140
63.2%56 / 140
0.0%100 / 132
0.0%100 / 132
69.7% best73 / 109
53.2%95 / 109
78.8% best67 / 87
72.2%73 / 87
77.2%42 / 73
77.3% best41 / 73