PulseAugur
中
实时 07:19:31
English(EN) RT @dylan__lim: Last Year We Built Megakernels. This Year We Deleted Them.

Together AI 从 megakernels 转向新的编译器以实现更快的 AI 推理

Together AI 宣布了其新的推理优化框架 Together Compiler,旨在提高 AI 模型性能。最初,该公司开发了将整个模型前向传播融合到单个 GPU 内核中的“megakernels”,实现了高速度但难以投入生产。通过进一步的实验,Together AI 发现通过利用 CUDA Graphs 和 Torch Inductor 等现有工具,他们可以在不编写自定义内核的情况下恢复大部分 megakernel 性能,甚至通过其新编译器超越之前的速度。 AI

影响 Together AI 的新编译器旨在优化推理性能,可能导致更快、更高效的 AI 模型部署。

排序理由 该集群描述了来自一家 AI 公司的新产品/框架发布,但不是前沿模型发布或重大的行业范围事件。

在 X — Together (inference / OSS) 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

Together AI 从 megakernels 转向新的编译器以实现更快的 AI 推理

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群描述了来自一家 AI 公司的新产品/框架发布,但不是前沿模型发布或重大的行业范围事件。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
infra, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
2 days old
Coverage has settled into its steady-state source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. X — Together (inference / OSS) TIER_1 English(EN) · togethercompute ·

    RT @dylan__lim: 去年我们构建了Megakernels。今年我们删除了它们。

    RT @dylan__lim: Last Year We Built Megakernels. This Year We Deleted Them. One year ago, we helped introduce some of the first AI megakern…

  2. Together AI blog TIER_1 English(EN) ·

    去年我们构建了巨型内核。今年我们删除了它们:通往 Together Compiler 之路

    Together Compiler uses AI agents to rewrite and fuse the GPU kernels in a model’s forward pass. It beats hand-written megakernels by 28% on B200 (2,118 vs 1,649 tokens/s on Llama 3.2 1B). This is how we got from one to the other.