🤯炸了！仅8GB显卡也能畅玩125B级Qwen3.8 Flash Next，速度稳定20+…
This is a dev post classified by AGK Intelligence as AI dev tools (a launch), kept by the Dev Radar because it carries real work, not commentary.
🤯炸了！仅8GB显卡也能畅玩125B级Qwen3.8 Flash Next，速度稳定20+ tok/s！支持128K上下文！多卡并行暴涨近100%，DFlash2轻量drafter让小显卡也能飞！ Strata v0.2.10正式发布，把本该需要几十上百GB显存的Qwen3.8-Flash-Next（125B MoE，每token仅激活6B）硬生生压进8GB显卡，还能稳住20+ tok/s！ 核心亮点 🔹8GB显卡就能跑：动态内存管理 + 专家卸载，128K上下文不再是梦，实测Intel Arc A750等卡片稳定出字。 🔹速度起飞：DFlash2替代MTP（仅约0.7GB），小显卡解码更快；双卡最高+98%（2×RTX A6000从201→399 tok/s），4×R9700 +78%。 🔹并行爆炸：All-Cards Split Engine默认开启，多请求吞吐大幅提升，长文档等待时间砍近一半。 🔹一键开箱：Windows/Linux一键安装，自动下载引擎和模型，直接提供本地OpenAI/Anthropic兼容API。 3步快速上手 1️⃣前往GitHub Releases下载安装包（Windows exe / Linux AppImage / 脚本一键安装）。 2️⃣启动Strata，自动识别显卡并下载对应引擎 + 推荐Qwen量化模型（ISTA-DASLab系
Posted by SuSu_酥酥👅 (25.9k followers) 1 h ago · 20 likes · 1.7k views · view the original post on X. Kept by the Dev Radar as AI dev tools. Tools mentioned: strata, qwen3.8-27b-longbook27-v2.
More dev work like this
- 🔥厉害了！DSH破甲核武器开源了！15步流水线一键生成定制破甲词，DeepSeek Harness直接起飞！ — @NFT_Chen
- your agents change. your memory stays. — @twigpine
- Meet this Turkish speech recognition model. It turns spoken Turkish into text using a… — @HuggingModels
- A new local-agent harness has a built-in BS detector. — @TeksEdge
- Scientific Agent Skill: A Library of Procedural Knowledge for — @gp_pulipaka
- Enterprise Agentic AI: From Strategy to Scale: An Executive Guide to Agents, RAG, Prompt… — @gp_pulipaka
- AI Prompt Engineering! #BigData #Analytics #DataScience #AI #MachineLearning #NLProc… — @gp_pulipaka
- Practical Agentic AI: Design, Build, and Deploy Autonomous LLM Powered Systems! #BigData… — @gp_pulipaka
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.5k posts from 5k X accounts over the last 21 days, 9.1k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-03 05:31 UTC. Full method.