AI Paper Atlas

Research map / Large language models / Agents, code and math

TritonGym: A Benchmark for Agentic LLM Workflows in Triton GPU Code Generation

Yue Guan, Yichen Lin, Xu Zhao, Jianzhu Yao, Xinwei Qiang, Zhongkai Yu, Pramod Viswanath, Yufei Ding, Adnan Aziz

ICML 2026 · Large language models › Agents, code and math · approach: Software engineering agents

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use AlphaEvolve workflow instead of one-shot generationClaude-4.5 · TritonGym full benchmarkPass@1: 31.6% -> 54.8%Improved
use AlphaEvolve workflow instead of one-shot generation (average over LLMs)LLM backbones · TritonGym full benchmarkPass@1: 32.8% -> 42.3%Improved
use Geak workflow instead of one-shot generation (average over LLMs)LLM backbones · TritonGym full benchmarkPass@1: 32.8% -> 36.7%Improved
use Geak workflow instead of one-shot generationClaude-4.5 · TritonGym full benchmarkPass@1: 31.6% -> 33.9%Mixed
use AlphaEvolve workflow instead of one-shot generationClaude-4.5 · TritonGym OOD samplesPass@1: 23.1% -> 61.5%Improved
use AlphaEvolve workflow instead of one-shot generationClaude-4.5 · TritonGym full benchmarkPerf@1: 0.307 -> 0.583Improved
use Geak workflow instead of one-shot generationGPT-5 · TritonGym full benchmarkPass@1: 23.9% -> 28.7%Improved
use Geak workflow instead of one-shot generationClaude-4.5 · TritonGym full benchmarkPerf@1: 0.307 -> 0.321Mixed
use Geak workflow instead of one-shot generationDeepSeek-V3 · TritonGym full benchmarkPass@1: 36.1% -> 38.1%Improved
use AlphaEvolve workflow instead of one-shot generationLlama-4 · TritonGym full benchmarkPass@1: 24.6% -> 35.1%Improved
use AlphaEvolve workflow instead of one-shot generationQwen-3 · TritonGym full benchmarkPerf@1: 0.282 -> 0.358Improved
use AlphaEvolve workflow instead of one-shot generationLLM backbones (average) · TritonGym DSL samplesPass@1: 14.6% -> 25.0%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy