询问一个问题，就是codex是否有办法以官方auth登录。然后不做custom。而是访问固定服务器中专。 不使用sub2api。而是通过比如插件等方式，让他访问的官方api地址改成服务器的。服务器只做中转。目的是避免ip变更等或者没开梯子导致暴漏真实IP。然后又能正常使用codex的所有功能。主要是考虑买个max20 。谁那有方案 1 个帖子 - 1 位参与者 阅读完整话题
Dear API customers, Our new coding model, grok-build-0.1 , is now available via the xAI API. Customers, including Notion and KiloCode , have loved its high speed, app building capabilities, and cost effectiveness. Try it now grok-build-0.1 was specifically trained for agentic coding tasks, including web development, debugging, and MCP support. It’s the same model that powers the Grok Build CLI . The model is served at a blazing-fast 100+ tokens/second and is priced at $1/m tokens in and $2/m tokens out. Outside of coding, it’s also a speedy, economical option for general-purpose agentic and tool calling use cases. To get started, create an API key here . grok-build-0.1 is also available via OpenRouter and Vercel AI Gateway . Sincerely, The xAI Team 1 个帖子 - 1 位参与者 阅读完整话题
限免随时可能结束，速度收。 App Store Widgetik: Custom Widgets App - App Store Download Widgetik: Custom Widgets by ANDREI RUSTI on the App Store. See screenshots, ratings and reviews, user tips, and more apps like Widgetik: Custom Widgets… 即时记录：只需轻点一下，即可更新您的饮水量、正念时长或习惯进度。 适配锁定屏幕：通过专门的锁定屏幕小组件（内联、矩形和圆形），让您最重要的信息随时可见。 满足各种需求的庞大小组件库 习惯追踪（交互式）：建立不可阻挡的日常习惯，目标进度一目了然。 计步器：与 Apple 健康 (Apple Health) 无缝同步，精美地记录您的每日进度。 正念与禅意：记录冥想和呼吸练习，让您全天保持专注。 活动倒计时：使用时尚的计时器为假期、周年纪念或重大发布倒计时间。 饮水追踪：通过快速点击记录饮水，保持健康与水分。 数字与指针时钟：从极简布局或经典表盘中随心选择。 习惯计数小组件：直接从主屏幕通过点击递增计数器来追踪日常习惯。 每日名言：每次解锁手机都能获得新鲜的动力，时刻保持灵感。 进度追踪：将您的周、月或年挑战可视化。 照片与图片小组件：将您最珍藏的回忆或审美灵感固定在主屏幕网格上。 月相与科技：实时追踪月相变化和电池状态。 1 个帖子 - 1 位参与者 阅读完整话题
chatgpt后端针对checkout_ui_mode只会返回custom，不管你请求参数里面的checkout_ui_mode了！虽然这只是一个账户的测试（也许是AB测试），但也反映了OpenAI很可能根据最近的举报已经采取行动！ 附图 脚本（没有加入优惠参数）: (async function checkoutLinkOnly() { try { const session = await fetch('/api/auth/session').then((r) => r.json()); const accessToken = session?.accessToken; if (!accessToken) { console.log('accessToken: null'); return; } const payload = { plan_name: 'chatgptplusplan', billing_details: { country: 'ID', currency: 'IDR', }, cancel_url: 'https://chatgpt.com/#pricing', checkout_ui_mode: 'hosted', }; const response = await fetch('https://chatgpt.com/backend-api/payments/checkout', { method: 'POST', headers: { Authorization: `Bearer ${accessToken}`, 'Content-Type': 'application/json', }, body: JSON.stringify(payload), }); const data = await response.json(); const hostedUrl = data?.url || data?.stripe_hosted_url || data?.checkout_url || null; console.log('accessToken:', accessToken); console.log('paymentLink:', hostedUrl); } catch (e) { console.log('accessToken: null'); console.log('paymentLink: null'); } })(); 8 个帖子 - 6 位参与者 阅读完整话题
codex桌面退出了之后，重新配置config之后桌面端咋没反应，cli没问题可以用 model_provider = “custom” model = “gpt-5.5” model_reasoning_effort = “xhigh” disable_response_storage = true approvals_reviewer = “user” service_tier = “fast” [model_providers.custom] name = “custom” wire_api = “responses” requires_openai_auth = true base_url = “ http://127.0.0.1:8317/v1 ” 4 个帖子 - 2 位参与者 阅读完整话题
API call failed (attempt 1/3): APIError 🔌 Provider: custom Model: gpt-5.5 🌐 Endpoint: https:······· 📝 Error: Our servers are currently overloaded. Please try again later. ⏱️ Elapsed: 1.98s Context: 6 msgs, ~9,077 tokens pro lite 美区正价订阅，我说怎么 hermes 老不干活。。 这该不会走 antigravity 的老路吧
API call failed (attempt 1/3): APIError 🔌 Provider: custom Model: gpt-5.5 🌐 Endpoint: https:······· 📝 Error: Our servers are currently overloaded. Please try again later. ⏱️ Elapsed: 1.98s Context: 6 msgs, ~9,077 tokens pro lite 美区正价订阅，我说怎么 hermes 老不干活。。 这该不会走 antigravity 的老路吧
Oracle Cloud Infrastructure Compute - Unavailable Virtual Machine Instance Oracle Cloud Infrastructure Customer, The host for the instance(s) listed in this notification experienced an issue, and we were unable to relocate your instance(s) to a new host due to a lack of availability of the specific shape in the Availability Domain your instance was deployed. We will continue to attempt to relocate your instance once the shape becomes available, but you can also take the following steps to restore access now: Resize to a different shape. See Changing the Shape of an Instance for more information. Terminate the instance and recreate it in a different Fault Domain or region. REF: COMPUTE-12D
Oracle Cloud Infrastructure Compute - Unavailable Virtual Machine Instance Oracle Cloud Infrastructure Customer, The host for the instance(s) listed in this notification experienced an issue, and we were unable to relocate your instance(s) to a new host due to a lack of availability of the specific shape in the Availability Domain your instance was deployed. We will continue to attempt to relocate your instance once the shape becomes available, but you can also take the following steps to restore access now: Resize to a different shape. See Changing the Shape of an Instance for more information. Terminate the instance and recreate it in a different Fault Domain or region. REF: COMPUTE-12D
Oracle Cloud Infrastructure Compute - Unavailable Virtual Machine Instance Oracle Cloud Infrastructure Customer, The host for the instance(s) listed in this notification experienced an issue, and we were unable to relocate your instance(s) to a new host due to a lack of availability of the specific shape in the Availability Domain your instance was deployed. We will continue to attempt to relocate your instance once the shape becomes available, but you can also take the following steps to restore access now: Resize to a different shape. See Changing the Shape of an Instance for more information. Terminate the instance and recreate it in a different Fault Domain or region. REF: COMPUTE-12D
Oracle Cloud Infrastructure Compute - Unavailable Virtual Machine Instance Oracle Cloud Infrastructure Customer, The host for the instance(s) listed in this notification experienced an issue, and we were unable to relocate your instance(s) to a new host due to a lack of availability of the specific shape in the Availability Domain your instance was deployed. We will continue to attempt to relocate your instance once the shape becomes available, but you can also take the following steps to restore access now: Resize to a different shape. See Changing the Shape of an Instance for more information. Terminate the instance and recreate it in a different Fault Domain or region. REF: COMPUTE-12D
Oracle Cloud Infrastructure Compute - Unavailable Virtual Machine Instance Oracle Cloud Infrastructure Customer, The host for the instance(s) listed in this notification experienced an issue, and we were unable to relocate your instance(s) to a new host due to a lack of availability of the specific shape in the Availability Domain your instance was deployed. We will continue to attempt to relocate your instance once the shape becomes available, but you can also take the following steps to restore access now: Resize to a different shape. See Changing the Shape of an Instance for more information. Terminate the instance and recreate it in a different Fault Domain or region. REF: COMPUTE-12D
claude desktop 不能用自定义的大模型了，有什么解决的办法吗？ message: Invalid custom3p enterprise config: inferenceModels: configured model “mimo-v2.5-pro” is not an Anthropic model. Gateway deployments require an Anthropic model from the provider catalog — expected a gateway model route referencing an Anthropic model (e.g. claude-sonnet-4-5, anthropic/claude-*). Name routes to match the underlying model. failingField: inferenceModels checkedAt: 2026-05-07T02:34:07.288Z 5 个帖子 - 4 位参与者 阅读完整话题
更新到最新版本，输入 /Pet 即可查看宠物 PS：如果需要custom电子坤坤，需要先去plugin安装hatch pet这个skill ，然后让全民制作人（Codex）制作就行 5 个帖子 - 3 位参与者 阅读完整话题
今天在docker里面装了hermes，然后cc也同步开启了hermes的配置，连接telegram，发现切换模型巨麻烦 3 个帖子 - 3 位参与者 阅读完整话题
Epoch AI Google controls the most AI computing power, driven by its custom TPUs We estimate Google is the largest single owner of AI compute, holding about one quarter of global cumulative capacity as of Q4 2025, primarily from its own custom TPU chips. 14 个帖子 - 14 位参与者 阅读完整话题
github.com/googleapis/python-genai docs: Add instruction for custom endpoint (#2335) main ← copybara/904525557 已合并 06:49PM - 23 Apr 26 UTC copybara-service[bot] +27 -0 docs: Add instruction for custom endpoint Copybara import of the project: -- 70 … d10f2c8e7068afd1439d3201f417ce725a3562 by AliyahZombie <[email protected]>: doc: Add instruction for custom endpoint FUTURE_COPYBARA_INTEGRATE_REVIEW=https://github.com/googleapis/python-genai/pull/1233 from AliyahZombie:patch-1 dbd593dcbb2b0a4a0b9c673914fd5181d641d745 去年的pr以为没后文了，昨天突然收到被合并的邮件！！ 也是混入Google的contributor了 1 个帖子 - 1 位参与者 阅读完整话题
custom_providers: name: cliproxy base_url: https://xxx/v1 key_env: OPENAI_API_KEY models: gpt-5.4: context_length: 131072 gpt-5.4-mini: context_length: 131072 gpt-5.3-codex: context_length: 131072 gpt-5.3-codex-spark: context_length: 131072 gpt-5.2: context_length: 131072 就自动配的，感觉不太行啊，我让它给我的最终回复加上一个@功能，给我干这么长时间 而且上下文长了好像还有幻觉，另一个任务怼着我上上上个消息在猛干 1 个帖子 - 1 位参与者 阅读完整话题