benchmarking
12 posts tagged “benchmarking”.
- Our Two Machines Ran Different Models Under the Same Name (2026-09-18)
- Speculative Decoding Won Alone and Lost Batched (2026-09-18)
- Your Headline tok/s Is a Short-Prompt Number (2026-09-18)
- We indexed 45 local-inference benchmark results — the 4.7x gap nobody looks for (2026-09-14)
- Our Benchmark Queues Deadlocked on Each Other's pgrep (2026-09-12)
- Our Kernel Benchmark Was Measuring the System Cache (2026-09-05)
- We Evaluated a Faster Inference Engine and Rejected It (2026-09-03)
- Benchmark the Hardware Before You Benchmark the Software (2026-09-02)
- Your SSH-Launched Benchmark Is Running at nice 5 (2026-09-02)
- Our Baseline Was the Three Worst Samples in the Log (2026-09-02)
- Every Surprising Eval Result Was Our Infrastructure (2026-08-15)
- The Benchmark Said 18%. Production Said 0.07%. (2026-08-10)