Lifei120

@AntLingAGI @OpenRouter 蚂蚁集团阿里巴巴

🎉 Good news — free access to Ling-3.0-flash on @OpenRouter has been extended until August 6 at 8:00 AM PT! ⚡ Try it for free: openrouter.ai/inclusionai/li… 💬 Questions or feedback? Join the Ant Ling Discord: discord.gg/GNaQc8WC5T

@yacineMTB 在cline dp is free

I ran out of codex credits. Should I try opencode

@yacineMTB use cline？

opencode sucks

@wangkezhangrich 陈行甲？

这些老登，为什么总喜欢吃饭的时候唱歌？

@0xSero deepseek 0731 因该排在最前面

Best models smallest to largest right now. - Gemma-4-12B - Qwen3.8-27B - Laguna-S2.1 - Deepseek-V4-Flash - Inkling-Small - MiniMax-M3 - GLM-5.* - Kimi-K2.7-Code - Qwen3.8-Max - Kimi-K3 Spoiled for choice, open weights community is much more exciting than the frontier.

@Seven_Nguyen666 听说你们越南人蛮团结

Có lẽ đây là bài viết khiến mình cảm xúc nhất từ khi build X. Gần 800 bình luận, gần 900 lượt thích, gần 100K view. Nhìn mấy con số này mới thấy những đêm thức khuya, những ngày ngồi nghĩ content cuối cùng cũng không uổng công. Cố lên anh em ơi. 💪

@MiaAI_lab in cline is free☺️

If you still don't know, DeepSeek v4 Flash is 90% off in Hermes through Nous Portal. Literally free.

@Web3Feng 62岁了

蔡崇信离婚的原因大概找到了。 网友评论：蔡崇信，310 老克勒 old money 律师世家出身，居然也还是喜欢整容的骚b

@kjckikyo 好白

@aijoey 有无审核的吗

I got MiniMax H3 running on a single DGX Spark 🎬 It took some experimenting with SM121 compatibility, online FP8, and a pinned vLLM Omni build, but the full text to video with audio workflow is now reproducible. I documented what worked, what broke, and the exact smoke tests here: Kudos to @Tech2Wild for showcasing it first. I just reverse engineered it. Video below took 2.5min to generate. So we have some work to do. cc: @vllm_project @MiniMax_AI @RyanLeeMiniMax github.com/joeynyc/MiniMa…

@louszbd 可以在z.ai用dp 吗

It's been great to hear from so many of you that ZCode feels good to use. Here are some updates. You can now: - Open and use web pages without leaving the app. - Preview PDFs and zoom in or out, better stay in flow when reading papers. - Open large files more reliably. - Use repo wiki to generate a catalog of architecture, with diagrams added if needed. v3.5.3 improves cache hit rate. Interactions are more responsive. Enjoy!

@opencode it free？

Qwen3.8 Max is now available in OpenCode Go

@kimuse_ @grok 韩国人好自由啊

@kimuse_ @grok 韩国有美国的军事基地吗、有几个，有多人人，他们工资是韩国人承担的吗

아직도 그렇게 높은 편은 아니지만 점점 가까워지고 호감도가 올라가는 이런 현상이 왜 일어나고 있는지 중국은 알아야한다. 역사적인 앙숙인 한국과 일본을 가깝게 만든 꽤 큰 이유 중에 하나는 중국.

@kimuse_ 不对，你是自由，连国家都在美爹的统治下，还自由，你说你们国家有几个美军基地

@elshayib_ 我也上瘾我看都上瘾了，可惜没钱本地部署fuck

I seriously think vibe coding is extremely addictive, and I'm not exaggerating at all, by looking at myself and others around me I see the same patterns you would in a proven addictive actions/substances. 1.Dopamine hits every time an agent finishes a task. 2.Looking for better stronger models. 3.Spending more gradually on subscriptions or hardware without real need for it sometimes. etc, I had more in mind but forgot about them while writing this 😅, but yeah am I the only one feeling that ?

@TheAhmadOsman 不幸的是，在中国我目前观察很少有人对本地ai感兴趣，我都热血沸腾了，我周围的同事都无动于衷，我都在x找到共鸣的，也许没钱本地部署吧，尤其kimi 3 ，minimax h3 ，dp 0731，qwen 3.8 max ，我觉得下个glm 也快来了期待吧，

I am genuinely so happy and proud of where Opensource and Local AI has gotten in these past few months. These past few months were so important in dictating how the future looks like and it has never looked brighter. We are all going to thrive together.

@TheAhmadOsman 是啊感谢中国🇨🇳

@cline 在cline 运行dp 0731好快啊，感谢cine

We are making the updated DeepSeek V4-Flash 0731 free in Cline. This is the first flash model we've found performs at SOTA levels, and are excited for you to feel the new frontier. 1. npm i -g cline 2. Open /settings > Cline provider 3. Select deepseek-v4-flash