一款名为 Saber 的新推理引擎已发布，使用 Rust 开发，专为 AMD R9700 GPU 设计。Saber 支持多种张量并行，特别支持 Qwen 3.8 27b 和 Flash next 模型。基准测试表明，Saber 在这些模型上的表现优于现有引擎，实现了高令牌生成速度。 AI
影响 为在 AMD 硬件上本地部署 LLM 提供了改进的推理速度。
排序理由 发布了针对特定硬件的新推理引擎。
AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →
新的 Saber 推理引擎为 AMD GPU 带来强劲性能
本文如何被排名
- Signal score
- 6 / 100
- Composite score across the factors below. Higher = stronger signal that this story matters right now.
- Newsworthiness bucket
- Tool
- 发布了针对特定硬件的新推理引擎。
- 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
- 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
- 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]
-
Qwen3.8 27b hitting 6,645 tok/s on prefill and 316 tok/s decode peak on Saber
<table> <tr><td> <a href="https://www.reddit.com/r/LocalLLaMA/comments/1xpicjz/qwen38_27b_hitting_6645_toks_on_prefill_and_316/"> <img alt="Qwen3.8 27b hitting 6,645 tok/s on prefill and 316 tok/s decode peak on Saber" src="https://preview.redd.it/jqp4n2gwx00i1.png?width=140&…