#benchmark

Posts on X2Q tagged benchmark. Updated as new posts are published.

• Jun 22, 2026

  Apple M1 Pro vs RTX 4070 Ti for local AI — GPU matmul and LLM benchmarks (2026)

  RTX 4070 Ti hits 75 TFLOPS at fp16 via tensor cores; M1 Pro MPS reaches 4.2 TFLOPS — an 18× compute gap. But on real LLM inference (minicpm-v4.6, qwen3.5, gemma4:12b via ollama), the RTX is only 3–4× faster. LLM inference at batch=1 is memory-bandwidth-bound, not compute-bound, and the M1's unified-memory architecture closes most of the gap. Apple's AMX extensions also give its CPU 2.7× better matmul throughput than an Intel i5 with the same thread count.

• Jun 22, 2026

  Rust vs Go vs C on Apple Silicon and Linux — benchmarks on real hardware (2026)

  Four benchmarks on two machines — Apple M1 Pro (macOS/clang) and Intel i5-13400F (Linux/gcc). Rust ≈ C on the M1; on x86 Linux, Rust is 1.5× slower than gcc on recursive fib and Go is 2.9× slower. Go's matrix multiply gap is 4.7× on Linux (vs 3.3× on arm64). The surprise: Rust's concurrent sum on x86 is 2.8× faster than C pthreads — LLVM likely folds the range into a closed-form sum.

• Jun 10, 2026

  Whisper backend shootout: faster-whisper vs HF Transformers vs whisper.cpp (2026)

  Three ways to run Whisper — faster-whisper (CTranslate2), Hugging Face Transformers (batched, SDPA), and whisper.cpp (ggml) — benchmarked on real Danish telephone audio on an RTX 4070 Ti. faster-whisper + large-v3-turbo is the fastest overall (~56× real-time); whisper.cpp CUDA is close at a fraction of the VRAM and even matches GPU speed on CPU for short files. Full speed (RTF), VRAM, and consensus-WER quality tables, plus a recommendation per use case.