to - 钛刻 - 科技风向旗 - 深度刻画技术趋势,引领数字未来 - 第76页 - 钛刻科技

Meta 与亚马逊 AWS 达成合作，部署数千万 Graviton 核心构建 AI 基础设施

IT之家 4 月 24 日消息，Meta 今天与亚马逊 AWS 达成合作协议，大规模部署 AWS Graviton 处理器，为下一代 AI 打下基础。据介绍，Meta 将首先部署数千万个 Graviton 核心，随后根据 AI 能力增长进度灵活扩展。虽然两家公司暂未披露具体合作

相关专题

tech www.ithome.com 2026-04-24 22:40:06+08:00

google pay买claude或gpt的订阅，会看卡的类型吗

如题，我如果用bybit的u卡已经绑定了google pay，然后我拿安卓手机在google store下载claude或者gpt进行订阅能够成功吗。另外如果封号了，那么这个费用是会把订阅的金额全额退回吗，如果在用了几天就封号了，那么还全额退了那不就是白嫖了几天嘛，没有这方面的经

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 22:23:10+08:00

API方式登录的Codex Desktop能使用插件功能吗

我发现主分支里的Superpowers是可以用插件的方式安装但是codex Desktop无法在API模式使用 3 个帖子 - 3 位参与者阅读完整话题

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 22:08:17+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 21:41:54+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 21:41:54+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 21:41:54+08:00

Claude Desktop支持用Chat2Api来调用模型吗?

因为最近看到Claude Desktop也支持第三方API了, 但是用Chat2Api又好像不是很行的样子佬们有教程的话求一个 3 个帖子 - 2 位参与者阅读完整话题

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 21:32:44+08:00

token出海, 说的就是deepseek吧?

今天看到deepseek v4的价格, 又想到原来看到的一个讨论: 普通人的翻身机会 = “token出海”. 感觉说的就是deepseek吧. 价格屠夫, 而且又不会被卡脖子. 锋哥牛逼!!! 1 个帖子 - 1 位参与者阅读完整话题

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 21:26:55+08:00

[求科普] codex+sub2api有副作用吗，比如contextcompact

多个codex账号的情况下，我现在用的是cockpit tools,一个账号额度不够了需要退出session再resume 我看站内其他帖子里很多佬有提到sub2api，我现在想转到sub2api或cliproxyapi 所以我想问一下能完美支持吗，会不会有副作用，比如conte

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 20:58:56+08:00

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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech www.v2ex.com 2026-04-24 20:35:40+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 20:35:40+08:00

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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech www.v2ex.com 2026-04-24 20:35:40+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 20:26:48+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 20:11:08+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 20:06:39+08:00

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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech www.v2ex.com 2026-04-24 19:59:44+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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech v2ex.com 2026-04-24 19:41:32+08:00

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

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech www.v2ex.com 2026-04-24 19:41:32+08:00

DeepSeekv4出预览版了，看看之前DeepSeek给我的回复。

https://linux.do/t/topic/1918604?u=lhllinuxdo 这是之前我问DeepSeek联网搜索时收到的回复，大家看看有多少是对的又有多少是错的呢？ 1 个帖子 - 1 位参与者阅读完整话题

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 19:17:43+08:00

一样的上下文, Claude占用Token是其他模型的1.5倍

以前在OpenCode中使用Claude和GPT的时候就发现. 同样GPT的聊天180ktoken, 一切到Claude就变成260ktoken, 我一度以为是GPT比较省Token. 现在用CC, 用Claude聊天, 显示160ktoken, 切到GLM5.1变成100kto

相关专题

Global Image Presentation Alert 专题内容 Advertising Comment Reporting Objective App Fashion Saving En...Data Plugin 专题内容 Forum Services Sport Objective 专题内容 Mobile Privacy Logo Recipe 专题内容 Backup Email Label Register Button Community Trading URL 专题内容 Saving Music Analytics Template Network Market 专题内容 Reporting 专题内容 Landing Engagement Navigation Upload Global Chapter Growth 专题内容 Image Revenue Saving Policy Seminar Workshop Tutorial Web 专题内容 Tcti 相关页面 Terms Tutorial Alert 专题内容 System File Guide Ranking Podcast Image 专题内容 Hosting Security 专题内容 Restaurant Faq Sales SEO Case Campaign Brand Section 专题内容 Security Demographic Digital API Brand Update Report Audience...Workshop Careers Guide Achievement Alliance Faq Feedback 专题内容 Strategy Planning Restaurant Tactic Quality Desktop 专题内容 File Economy Dashboard Conference Reporting Accessibility Pri...Calendar Login 专题内容

tech linux.do 2026-04-24 19:04:52+08:00

to - 钛刻 - 科技风向旗 - 深度刻画技术趋势,引领数字未来 - 第76页 - 钛刻科技 | TCTI.cn

相关标签