PulseAugur
中
实时 11:10:33
English(EN) Step 3.7 Flash: The 198B MoE Model Everyone Is Actually Running

StepFun 的 198B MoE 模型以低成本提供专业级性能

StepFun 发布了 Step-3.7-Flash,这是一个拥有 1980 亿参数的混合专家(MoE)模型,每个 token 仅使用 110 亿激活参数进行推理。这种架构实现了高吞吐量,每秒可达 400 个 token,并显著降低了计算成本。该模型包含一个用于图像理解的视觉编码器,并在各种基准测试中,特别是在编码和视觉问答任务上,展现出与 GPT-5.5 和 Claude Opus-4.6 等领先模型相媲美的性能。 AI

影响 该模型具有成本效益的推理能力,可能会加速大型、强大模型在生产环境中的应用。

排序理由 前沿实验室(StepFun)发布新模型,包含详细的技术规格和基准测试对比。[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 dev.to — LLM tag 阅读 →

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

StepFun 的 198B MoE 模型以低成本提供专业级性能

本文如何被排名

Signal score
49 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
前沿实验室(StepFun)发布新模型,包含详细的技术规格和基准测试对比。[lever_c_demoted from frontier_release: 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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. dev.to — LLM tag TIER_1 English(EN) · Dishant Sharma ·

    Step 3.7 Flash:大家实际运行的 198B MoE 模型

    <p>someone on X posted a photo of a DGX Spark sitting on a regular desk with a terminal window running step 3.7 flash. a 198 billion parameter vision model. on a box that fits next to a monitor. and it was not a flex post. it was a "here is the config that saved me three hours" p…