慎为

AI Application Architecture

[慎为 @actwithcaution]

@GOJO_6E 如果是真的就太棒了

[rLLM @rllm_project]

1/5 New rLLM Blog Post: Continual learning for agents 🧵

Can an AI agent keep learning after deployment—from real production traffic?

Each interaction yields one trajectory. Unlike build-time RL, you typically can't sample multiple rollouts for the same prompt, so GRPO's group-relative advantage no longer applies.

Our result shows a +16.2 points improvement on real SWE tasks.

Full post: https://t.co/P6Sog0P3Tg

[Jay Chooi @chooi_jeq]

GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-like figure, heat compressed gas, or produce toxic fumes, succeeding in 62% of its attempts. Fable 5.1 refused more often, attempting 80% of trials and completing 34%.

[行者AI视频 @joshesye]

用 AI 就能把色卡换装视频做得一模一样，真没那么难。

使用Hypit，在codex 里直接就能一比一复刻出来。

天凉了，正好让模特换上秋装。创意来自 @原来是陶阿狗君

[DΞV @junwatu]

Qwen Image 2.1 for character style infographics.

I didn't expect it at first, but after looking through these generated images, this model might actually also great at UI design.

Prompt (first image) ⬇️

[慎为 @actwithcaution]

@marukoro1983 祝你生日快乐，万事顺心。

[慎为 @actwithcaution]

@tianzige4 温暖又感人

[Enze Xie @xieenze_jr]

🚀 Sol-H3 on DGX Spark: 768p in Under a Minute 🤩

Monday: 8×B300, 5s 768p in 1.65s — faster than playback.
Today: the same stack on one desktop Spark — about 56s hot E2E.

Five seconds of 1344×768 video at 24 FPS with stereo audio, on a single NVIDIA DGX Spark (GB10).

Two-stage, not the datacenter profile:
384p H3 draft → latent ×2 → H3-to-LTX VAE adapter → 768p LTX refine → VAE decode

No decode/re-encode between stages. Stage 2 is conditioned on the draft latent, so Gemma stays off the box. Quantized weights stay resident; sparse attention cuts the rest.

Stage 1 takes any MiniMax-H3 few-step LoRA. Timing is hot E2E (encode → both stages → video/audio VAE); cold start and MP4 mux are separate. Apache 2.0.

Server was realtime. Edge is one box, under a minute.

🔗 https://t.co/rfo0EVkJsI

Amazing team effort—full credits in the blog.
@haopengl33 @lawrence_cjs @yitongli165665 @shanasaimoe ,Jingyu Xin, @HaochengXiUCB @songhan_mit

[E ★ @reminiscinbby]

Dolomites, Italy

[慎为 @actwithcaution]

@jenwxie 我喜欢《阿卡兹班的囚徒》那部。

[慎为 @actwithcaution]

@ilovek8s 牛🐂🐂

[Yanjie Ze @ZeYanjie]

GPT6 Astra solved Rubik’s Cube with robot hands

https://t.co/Hkh1hi1VRz

[Higgsfield AI 🧩 @higgsfield_ai]

This shot walks through seven rooms without a cut.

GPT-6 Astra mapped the space, the blocking never left the viewport, and Higgsfield Seedance 2.5 rendered the whole run.

[慎为 @actwithcaution]

@sofiatieger 哈哈哈哈，我说怎么每次都找不到插孔

[慎为 @actwithcaution]

@Umesh__digital 这样都好几年了，兄弟，我快扛不住了

[慎为 @actwithcaution]

@ashebytes 太酷了

[慎为 @actwithcaution]

@TheBestqueenx 哈哈哈，你行你上

[Charly Wargnier ♨️ @DataChaz]

normal coder vs vibe-coder 😭

[慎为 @actwithcaution]

@DataChaz