SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.09708 · 2026

Metal-Sci: A Scientific Compute Benchmark for Evolutionary LLM Kernel Search on Apple Silicon

Víctor Gallego

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 3 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2605.09708.json · badge: /api/ran/2605.09708.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
vicgalle/metal-sci-kernels canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
FunctionStatusWhere it lives
call_llm Ran
ran; our contract draft was wrong, not the code
vicgalle/metal-sci-kernels/metal_kernels/llm.py
pointer only (licence: NONE) · get_code("f740470510a7fa9e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

detect_stagnation Ran
ran, violated the contract we drafted
vicgalle/metal-sci-kernels/metal_kernels/prompts.py
pointer only (licence: NONE) · get_code("4902382cc86f379a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

gb_per_s Ran
ran, honoured the contract we drafted
vicgalle/metal-sci-kernels/metal_kernels/task.py
pointer only (licence: NONE) · get_code("ead580fb3bc849a6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

gflops Ran
ran, honoured the contract we drafted
vicgalle/metal-sci-kernels/metal_kernels/task.py
pointer only (licence: NONE) · get_code("f6060a15f03f7f7a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

is_gemini Ran
ran, violated the contract we drafted
vicgalle/metal-sci-kernels/metal_kernels/llm.py
pointer only (licence: NONE) · get_code("6a4038fbd86d47bd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

is_openai Ran
ran, violated the contract we drafted
vicgalle/metal-sci-kernels/metal_kernels/llm.py
pointer only (licence: NONE) · get_code("f8137c8464e14160")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

register_task Ran
ran; our contract draft was wrong, not the code
vicgalle/metal-sci-kernels/metal_kernels/task.py
pointer only (licence: NONE) · get_code("c827ccdf941d207c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, vicgalle/metal-sci-kernels when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
call_llm 1 other paper
1 later
2608.08722 (2026-08)
Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure
is_gemini 1 other paper
1 later
2608.08722 (2026-08)
Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure
is_openai 1 other paper
1 later
2608.08722 (2026-08)
Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present METAL-SCI, a 10-task benchmark of scientific Apple Silicon Metal compute kernels spanning six optimization regimes (stencils, all-pairs in n-body problems, multi-field Boltzmann, neighbor-list molecular dynamics, multikernel PDE, FFT). Each task ships a CPU reference, a roofline-anchored fitness function, and a held-out generalization size. We pair the benchmark with a lightweight harness for automatic kernel search that runtime-compiles each candidate, scores it against the roofline across multiple sizes, and feeds structured compile and per-size correctness diagnostics back to a frozen LLM driving a (1+1) evolutionary loop. We report matched single-model sweeps of Claude Opus 4.7, Gemini 3.1 Pro, and GPT-5.5 on M1 Pro: indistribution self-speedups span 1.00× to 10.7×. Beyond raw speedup, our central methodological claim is structural: the held-out gate scoring function Φ T (evaluated once at end-of-run on a configuration the agent never sees during search) functions as a cheap mechanical oversight primitive on this automatic search loop, catching e.g. an Opus template <uint D> HMC win that returns wrong samples at unseen dimensions, and a GPT FFT3D best that wins in-distribution at 2.95× speedup but collapses to 0.23× on a 256 3 held-out cube, a silent regression that the in-distribution score alone cannot see. Code at github.com/vicgalle/metal-sci-kernels.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2605.09708")
get_code_for_paper("2605.09708")
have("2605.09708")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2605.09708.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2605.09708.svg)](https://syntology.ai/paper/2605.09708)

Connect an agent — have() is free.