PulseAugur
中
实时 07:27:30
English(EN) I Tested SWE-1.6 on 18 Coding Tasks — Cognition Killed SWE-1.5 With Just Post-Training

Cognition 的 SWE-1.6 模型在编码任务上取得重大进展

最近对 Cognition 的 SWE-1.6 模型在 18 项编码任务上的评估显示,其相比前代 SWE-1.5 有了显著改进。新版本在性能上比 Cognition 此前的旗舰模型提高了 10 个百分点。值得注意的是,SWE-1.6 以更少的对话轮次完成了这一目标,并保持了每秒 950 个 token 的相同处理速度。 AI

影响 在编码任务中展示了显著的性能提升,可能影响未来 AI 编码助手的开发。

排序理由 该集群报告了对特定 AI 模型在编码任务上性能的评估,属于研究范畴。[lever_c_demoted from research: ic=1 ai=1.0]

在 Towards AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

Cognition 的 SWE-1.6 模型在编码任务上取得重大进展

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
该集群报告了对特定 AI 模型在编码任务上性能的评估,属于研究范畴。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, 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
193 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. Towards AI TIER_1 English(EN) · Chew Loong Nian - AI ENGINEER ·

    我用18个编码任务测试了SWE-1.6 — 仅凭训练后认知能力就超越了SWE-1.5

    <div class="medium-feed-item"><p class="medium-feed-image"><a href="https://pub.towardsai.net/i-tested-swe-1-6-on-18-coding-tasks-cognition-killed-swe-1-5-with-just-post-training-5ae8eb187d8d?source=rss----98111c9905da---4"><img src="https://cdn-images-1.medium.com/max/1672/1*Nb_…