MetalBench – Benchmark for Apple Silicon's Metal Shading Lang
MetalBench is a benchmarking tool designed for Apple Silicon's Metal Shading Language. It allows users to evaluate the performance of Metal GPU kernels against MLX reference implementations. The tool features an agent-based system that automates kernel profiling, optimization, and verification to enhance performance.
- ▪MetalBench benchmarks Apple Metal GPU kernels against MLX reference implementations.
- ▪The tool includes a closed-loop agent system for profiling and optimizing kernels.
- ▪It supports various kernel sets, including basic operations and full model blocks.
Hacker News (Newest) files mainly under programming. We currently carry 5,306 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/Lazarus-931/MetalBench |
| Publication time | Mon, 25 May 2026 03:03:24 +0000 |
| Retrieval time | 2026-05-25T03:37:35.249Z |
| Last seen | 2026-05-25T03:37:35.249Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | fiW0TnszAwb4 |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
MetalBench - Metal-based kernel benchmark, library, and closed-loop agent authoring system Benchmarking Apple Metal GPU kernels against MLX reference implementations. Modeled on KernelBench, swapping CUDA → Metal and PyTorch → MLX. While working on inference for Apple Silicon, I found an agent-based loop of kernel writing & testing against a perf/accuracy test helped a lot along the way. This repo contains the harness and code used to benchmark against baseline MLX versions. Kernels don't differ much — mostly in how threadgroups are utilized. One of the main differences was performance across the newer M-series chips.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.