Compare
Compare local AI hardware with workload-aware output.
RTX 4090 24GB wins for coding in balanced mode
Based on model fit, speed, and quality across top recommendations.
MacBook Pro M2 Pro 16GB
ACodeGeeX 4 9B
llama.cppq4-k-mRuns well
8.7 GB / 11.5 GB
27.9 tok/s89K ctx
AGemma 4 E4B
llama.cppq4-k-mRuns well
8.8 GB / 11.5 GB
23.4 tok/s50K ctx
SQwen 3.5 9B
llama.cppq4-k-mTight fit
10.3 GB / 11.5 GB
27.4 tok/s25K ctx
Quick comparison
| Metric | MacBook Pro M2 Pro 16GB | RTX 4090 24GB |
|---|---|---|
| Models that fit | 3 | 3 |
| Avg decode tok/s | 26.2 | 37.9 |
| Best grade score | 92 | 93 |