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.
GTX 1080 Ti 11GB
ACodeGeeX 4 9B
llama.cppq4-k-mRuns well
8.1 GB / 11.0 GB
56.9 tok/s92K ctx
AGemma 4 E4B
llama.cppq4-k-mRuns well
8.2 GB / 11.0 GB
47.7 tok/s51K ctx
SQwen 3.5 9B
llama.cppq4-k-mTight fit
9.7 GB / 11.0 GB
55.9 tok/s26K ctx
Quick comparison
| Metric | GTX 1080 Ti 11GB | RTX 4090 24GB |
|---|---|---|
| Models that fit | 3 | 3 |
| Avg decode tok/s | 53.5 | 37.9 |
| Best grade score | 94 | 93 |