English(EN) Spark X2.5-4B & 1.7B: the only on-device models with native 1M-token context — now open source

SparkLLM发布具有1M token上下文的开源设备端模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 03:21

SparkLLM发布了两款开源的设备端语言模型Spark X2.5-4B和Spark X2.5-1.7B，均具备原生100万token的上下文窗口。这种扩展的上下文能力使模型能够处理和推理大量信息，例如整个代码库或长篇文档，而无需分块。这些模型在代码生成、办公文档分析、智能家居控制和机器人等任务中表现出色，可与大型云端模型相媲美。 AI

影响支持更复杂的设备端推理和任务执行，可能减少在需要广泛上下文的任务中对云端模型的依赖。

排序理由来自一家拥有重要能力（设备端1M上下文窗口）的实验室的新开源模型发布。[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 →

本文如何被排名

Signal score: 61 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Significant; 来自一家拥有重要能力（设备端1M上下文窗口）的实验室的新开源模型发布。[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]

dev.to — LLM tag TIER_1 English(EN) · SparkLLM · 2026-09-01 03:21

Spark X2.5-4B & 1.7B：唯一拥有原生1M token上下文的设备端模型——现已开源

Today SparkLLM releases and open-sources two on-device general models: Spark X2.5-4B and Spark X2.5-1.7B. Both natively support a context window of up to 1,000,000 tokens — as far as we know, the only on-device models to do so…

本文如何被排名

报道来源 [1]

Spark X2.5-4B & 1.7B：唯一拥有原生1M token上下文的设备端模型——现已开源

相关实体

相关话题