Unified MemoryBenchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Models
MiniCPM-V 4.7 1B
openbmb/MiniCPM-V-4.7-1B-gguf:Q4_K_M · llama.cpp · llama-server version: 0.7.0 (build 11669, commit 620d2225c)
#3 of 3
reading images
72.5
quality score
199.5
median tokens/s
2.09 GB
RAM with 8K context filled
1.47 GB
download size
Score per job
| Job | Accuracy | Pass | Easy | Medium | Hard | Median latency | Tok/s | Job time | Errors |
|---|---|---|---|---|---|---|---|---|---|
| Read images | 72.5 | 72.5% | 100.0% | 66.7% | 53.8% | 1.19 s | 199.5 | 45.0 s | 0 |
Easy / medium / hard columns are pass rates within each difficulty tier.
Memory and loading
| Download size | 1.47 GB |
| Loaded, idle | 1.97 GB |
| In use (8K context) | 2.04 GB |
| 8K context filled | 2.09 GB |
| Server peak during the jobs (incl. reserved context) | 5.18 GB |
| Load time | 32.3 s |
Measured on Apple M4 Max, 36 GB, macOS 27.0.1. How memory was measured.