🤖 MiniCPM5-1B (Q8_0) · CPU
100k context · Max output 30k · Hybrid Reasoning ON · OpenAI API at
/v1/chat/completions
Explain transformer attention in 3 sentences
Write a Python dedup function
Three habits for senior DevOps
Draft a bug-fix release note
Send
⚙️ Settings
System prompt
Temperature
0.70
Top-p
0.90
Max tokens
4096
Repeat penalty
1.05
📊 Live Metrics
Status
idle
TTFT
—
Reasoning tok/s
—
Response tok/s
—
Total tok/s
—
Reasoning count
0
Response count
0
Total tokens
0
Elapsed
—
Gen time
—
Total time
—