Llama 3.1 8B vs MiniCPM5-1B
Change modelsLlama 3.1 8BMeta
MiniCPM5-1BOpenBMB
IndexAll categories, out of 100.
17.9 ±9.6 best
15.5 ±7.0
GapIndex points below the top.
Top
−2.4 · no clear gap
CoverageIndex weight that has results.
55%
75% best
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2024-07
2026-05
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
131K best
AgenticMulti-step tasks with tools.
—
8.5
CodingCode writing and repair.
13.9 best
11.9
ReasoningLogic problems and puzzles.
—
—
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
14.4 best
9.2
MultilingualTasks in many languages.
24.8 best
5.0
InstructionTasks with strict prompt rules.
25.1
28.9 best
MathMath problems.
10.0
21.9 best
Benchmarks
5 sharedOnly Llama 3.1 8B: 15Only MiniCPM5-1B: 13
Llama 3.1 8B
MiniCPM5-1B