Unified MemoryBenchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Models
MiMo-V2.6 Distill 9B
bartowski/MiMo-V2.6-Distill-Qwen-9B-GGUF:Q4_K_M · llama.cpp · llama-server version: 0.7.0 (build 11669, commit 620d2225c)
#1 of 8
overall, text jobs
#2 of 3
reading images
91.0
quality score
37.8
median tokens/s
8.03 GB
RAM with 8K context filled
6.30 GB
download size
Score per job
| Job | Accuracy | Pass | Easy | Medium | Hard | Median latency | Tok/s | Job time | Errors |
|---|---|---|---|---|---|---|---|---|---|
| Assistant actions | 86.0 | 84.0% | 100.0% | 82.4% | 72.2% | 1.35 s | 37.3 | 1m 19s | 0 |
| Ask my documents | 90.8 | 90.0% | 100.0% | 92.3% | 80.0% | 4.11 s | 40.2 | 2m 48s | 0 |
| Rewrite and proofread | 90.0 | 64.3% | 92.9% | 78.6% | 21.4% | 3.12 s | 36.2 | 2m 23s | 0 |
| Summarize + extract | 96.1 | 75.0% | 80.0% | 76.9% | 68.8% | 3.62 s | 37.6 | 2m 55s | 0 |
| Translate | 89.5 | 72.5% | 84.6% | 69.2% | 64.3% | 2.05 s | 37.8 | 2m 09s | 0 |
| Sort and triage | 93.8 | 92.9% | 94.1% | 100.0% | 85.0% | 0.72 s | 43.4 | 38.6 s | 0 |
| Read images | 77.5 | 77.5% | 91.7% | 73.3% | 69.2% | 1.99 s | 43.1 | 1m 26s | 0 |
Easy / medium / hard columns are pass rates within each difficulty tier.
Memory and loading
| Download size | 6.30 GB |
| Loaded, idle | 7.62 GB |
| In use (8K context) | 7.82 GB |
| 8K context filled | 8.03 GB |
| Server peak during the jobs (incl. reserved context) | 18.28 GB |
| Load time | 6.10 s |
Measured on Apple M4 Max, 36 GB, macOS 27.0.1. How memory was measured.