Devstral Small vs Llama 3.3 70B
Change modelsDevstral SmallMistral
Llama 3.3 70BMeta
IndexAll categories, out of 100.
28.2 ±8.2
28.2 ±5.3
GapIndex points below the top.
Top
Top
CoverageIndex weight that has results.
75%
90% best
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2025-05
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.
21.6
27.2 best
CodingCode writing and repair.
35.7 best
28.9
ReasoningLogic problems and puzzles.
28.5 best
27.5
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
24.2
29.1 best
MultilingualTasks in many languages.
—
34.1
InstructionTasks with strict prompt rules.
21.4
40.9 best
MathMath problems.
—
15.6
Benchmarks
6 sharedOnly Devstral Small: 2Only Llama 3.3 70B: 10
Devstral Small
Llama 3.3 70B