Benchmark comparison on MacBook Pro M3 Max 128GB: Splash, Ollama (MLX), and Ollama (llama.cpp) across Qwen3.6-35B and Qwen3.8-27B
benchmark
Fields
id
sources/src-incoai-splash-issue-37Publisher
incoai / community
Kind
benchmark
Published
2026-09-30
Accessed
2026-10-11
notes
Direct head-to-head empirical measurement on MacBook Pro M3 Max (16C CPU / 40C GPU, 128 GB unified memory at 400 GB/s) comparing MLX and llama.cpp runtimes under Ollama on the exact same machine and hardware state.
Notes
Direct head-to-head empirical measurement on MacBook Pro M3 Max (16C CPU / 40C GPU, 128 GB unified memory at 400 GB/s) comparing MLX and llama.cpp runtimes under Ollama on the exact same machine and hardware state.
Cross-references
This record references no other records.