Al - 钛刻 - 科技风向旗 - 深度刻画技术趋势,引领数字未来 - 第29页 - 钛刻科技

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

tech v2ex.com 2026-04-25 03:39:55+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-25 03:39:55+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-25 02:39:55+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-25 02:18:07+08:00

【深夜福利】免费送 GLM Coding Plan 体验卡 - 自取

剩余 3/3 张 https://bigmodel.cn/activity/trial-card/1BMR381KZD 3 个帖子 - 3 位参与者阅读完整话题

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech linux.do 2026-04-25 02:08:51+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-25 01:40:43+08:00

CPA 内网ip访问 webUI 登录失败: 访问被拒绝，权限不足

docker部署,官网教程部署成功 localhost访问没问题内网ip访问则报标题错误(无论部署机还是其他机) 配置方面: 确认密码没错确认开启了 allow-remote host为 0.0.0.0 localhost进行活动,docker能看到访问日志. ip访问日志都

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech linux.do 2026-04-25 00:57:46+08:00

被标记为网络安全风险有啥影响不好怕被封号

此聊天已被标记为可能存在网络安全风险背景：我想codex帮我分析 GitHub - TheMythologist/GenP: Automatically builds GenP executables from source · GitHub 这个项目结果直接报错让我加入什么

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech linux.do 2026-04-25 00:29:31+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:54:14+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:54:14+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:54:14+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:47:32+08:00

有用axonhub的佬吗，目前能转发gpt image 2吗？

我配上去后拿cherry studio测试报错：failed to transform request: invalid request: image is not supported 不知道是我配置问题，还是暂时还不支持 1 个帖子 - 1 位参与者阅读完整话题

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech linux.do 2026-04-24 23:33:36+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:03:46+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 23:03:46+08:00

[Visual Studio Code] 有没有办法在 vscode 中的 ai 插件对话中做标注？

比如高亮，能够让我快速定位到标注过的地方就行，比搜索会方便很多

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 22:55:07+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 22:48:43+08:00

[Local LLM] 我做了个工具让 8GB 显卡跑 30B 模型从 3 tok/s 提到 21 tok/s，记录一下技术发现

最近在折腾本地大模型，发现一个核心问题：Ollama 和 LM Studio 能让模型跑起来，但参数全靠猜——上下文长度、KV cache 类型、MoE expert 放哪、ubatch 多大……用默认参数基本是在浪费显卡。于是做了个工具自动找最优配置，过程中踩了不少坑，记录一

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 22:41:38+08:00

[分享创造] 分享一个我做的 GPT Image 2 prompt gallery，整理了 400+ 提示词和模板

大家好，最近我做了一个 GPT Image 2 prompt gallery： https://gptimagelab.com 同时把一部分内容整理成了 GitHub 仓库： https://github.com/peterRooo/awesome-gpt-image-2-pro

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech v2ex.com 2026-04-24 22:06:05+08:00

realme 真我商城明起停止运营，业务全面整合至 OPPO 体系

IT之家 4 月 24 日消息，据 realme 真我商城官网公告，真我商城将于 4 月 25 日（明天）停止提供购物、商品内容浏览、用户反馈、活动参与、商品评价、帖子评论、观看视频与直播、积分服务，保留历史订单信息查询服务等。“真我商城”小程序也将同步下架。对于有查看历史订单

相关专题

Shopping Recipe Traffic Subject AI Experience Link Mobile Les...Campaign 专题内容 Share Beauty Reporting Settings Objective 专题内容 Behavior Retention Discount Mobile 专题内容 Collaboration Solution Careers Landing Productivity Study Tea...Email Whitepaper Recommendation Logo Media 专题内容 Kpi Server Hosting Premium Cost 专题内容 Partner Forecast AI Rating Keyword Lead Support Reminder 专题内容 Form Alert Machine Forecast 专题内容 Campaign Training 视频 Shopping Cost 专题内容 Reminder Backup 专题内容 Hosting Settings Tool Web 游戏 Research Entertainment Calculato...Ebook Team Discovery Tutorial Engagement Lead Vacation 专题内容 Travel Global Value Visitor Management Report App User Recomm...Security Efficiency Analytics Partner Account 专题内容 Growth 专题内容 Restore Global Target Roi Schedule Website 专题内容 Client Whitepaper Customer 专题内容 Template Privacy Hotel Vacation 专题内容 Website Logo Update Sport Conference 游戏 Budget 专题内容

tech www.ithome.com 2026-04-24 21:50:19+08:00

Al - 钛刻 - 科技风向旗 - 深度刻画技术趋势,引领数字未来 - 第29页 - 钛刻科技 | TCTI.cn

相关标签