# KernelBench leaderboard

Coding benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/coding-kernel-bench

KernelBench Hard H100. Every model the catalog carries a published KernelBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Coding |
| Measure | Mean peak fraction of hardware roofline over valid cells |
| Tasks | 6 GPU-kernel optimization problems |
| Difficulty | Agentic GPU systems engineering |
| Published by | [KernelBench Hard](https://kernelbench.com/hard?gpu=h100) |

An agentic GPU-kernel benchmark that measures how much of the hardware roofline a model's correct, audit-clean kernels reach on six demanding CUDA and Triton problems.

## KernelBench ranking

No model in the catalog has a published KernelBench score. The benchmark is defined by KernelBench Hard, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
