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.
Intel Arc Pro A60 12GB
SQwen 3.5 9B
llama.cppq4-k-mRuns well
9.8 GB / 12.0 GB
36.8 tok/s32K ctx
ACodeGeeX 4 9B
llama.cppq4-k-mRuns well
8.2 GB / 12.0 GB
37.5 tok/s116K ctx
AGemma 4 E4B
llama.cppq4-k-mRuns well
8.3 GB / 12.0 GB
31.4 tok/s63K ctx
Quick comparison
| Metric | Intel Arc Pro A60 12GB | RTX 4090 24GB |
|---|---|---|
| Models that fit | 3 | 3 |
| Avg decode tok/s | 35.2 | 37.9 |
| Best grade score | 96 | 93 |