Llama 3.3 70B vs Llama 4 Maverick
Change modelsLlama 3.3 70BMeta
Llama 4 MaverickMeta
IndexAll categories, out of 100.
28.2 ±5.3
34.4 ±4.7 best
GapIndex points below the top.
−6.2 · no clear gap
Top
CoverageIndex weight that has results.
90%
95% best
SpeedOutput tokens per second.
—
61/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2024-12
2025-04
Input / 1MUS dollars per 1M input tokens.
—
$0.35
Output / 1MUS dollars per 1M output tokens.
—
$1.00
1K chats2K tokens in, 500 out, each.
—
$1.20
1K documents20K tokens in, 1K out, each.
—
$8.00
1K agent tasks200K tokens in, 20K out, each.
—
$90.00
ContextMaximum tokens in one request.
128K
524K best
AgenticMulti-step tasks with tools.
27.2
36.9 best
CodingCode writing and repair.
28.9
37.2 best
ReasoningLogic problems and puzzles.
27.5
36.7 best
MultimodalTasks with images and text.
—
36.0
KnowledgeFacts and expert knowledge.
29.1
34.0 best
MultilingualTasks in many languages.
34.1
—
InstructionTasks with strict prompt rules.
40.9 best
29.2
MathMath problems.
15.6
28.6 best
Benchmarks
10 sharedOnly Llama 3.3 70B: 6Only Llama 4 Maverick: 9
Llama 3.3 70B
Llama 4 Maverick