Gemma 4 E4B vs Llama 3.3 70B
Change modelsGemma 4 E4BGoogle
Llama 3.3 70BMeta
IndexAll categories, out of 100.
28.2 ±5.2
28.2 ±5.3
GapIndex points below the top.
Top
Top
CoverageIndex weight that has results.
90%
90%
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-04
2024-12
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.
128K
128K
AgenticMulti-step tasks with tools.
25.3
27.2 best
CodingCode writing and repair.
27.4
28.9 best
ReasoningLogic problems and puzzles.
30.9 best
27.5
MultimodalTasks with images and text.
32.8
—
KnowledgeFacts and expert knowledge.
30.2 best
29.1
MultilingualTasks in many languages.
8.8
34.1 best
InstructionTasks with strict prompt rules.
30.3
40.9 best
MathMath problems.
—
15.6
Benchmarks
9 sharedOnly Gemma 4 E4B: 15Only Llama 3.3 70B: 7
Gemma 4 E4B
Llama 3.3 70B