Compare

Compare local AI hardware with workload-aware output.

Operating mode: Balanced. Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

RTX 4090 24GB wins for coding in balanced mode

Based on model fit, speed, and quality across top recommendations.

SQwen 3.5 4B

llama.cppQ4_K_MRuns well

6.3 GB / 8.0 GB

48.0 tok/s28K ctx

AGemma 4 E2B

llama.cppQ4_K_MRuns well

5.3 GB / 8.0 GB

61.2 tok/s96K ctx

ACodestral Mamba 7B

llama.cppQ4_K_MRuns well

6.5 GB / 8.0 GB

84.0 tok/s67K ctx

Winner

SDevstral Small 2 24B Instruct

llama.cppQ4_K_MTight fit

20.4 GB / 24.0 GB

40.0 tok/s40K ctx

SCodestral 2 25.08

llama.cppQ4_K_MRuns well

19.2 GB / 24.0 GB

41.7 tok/s48K ctx

SQwen 3.6 27B

llama.cppQ4_K_MTight fit

20.7 GB / 24.0 GB

20.2 tok/s69K ctx

Quick comparison