Project.I Cpt I - Get Familiar with Diffusers

I wanna doooominate CivitAI. Since this probably requires large-scale training on a server that possess multiple GPUs, Diffusers would be an invincible choice. My local GPU got only 16GB VRAM which is intensive for finetuning the whole SDXL network, I decided to train an LoRA simply for gathering practical experience. I won’t rent a server until everything, including my diffusers techniques, is ready. Code I mentioned can be found in the repo below....

July 21, 2025 · 2 min · A1yCE

Development Dairy1

Recruiting AFlodit as 3D Modeller, Lighthouse (actually not only this blog, but also our team’s name) is going to reach its great advent. With the recent establishment of our task tracker, each team member is now clearly aware of their responsibilities, giving us the feeling of becoming increasingly professional and efficient. AFlodit primarily uses 3DMax, while ty and I am accustomed to Blender. Although we initially had concerns regarding potential workflow inconsistencies, our main operations will focus on Unreal Engine’s subsystems, such as Nanite and Landscape Editor, so theoretically the choice of different modelling software won’t make a difference....

March 20, 2025 · 2 min · A1yCE

Faraway

During my middle school years, I composed an essay about my grandmother residing in the countryside, which moved my Chinese teacher profoundly. However, at that time, I did not perceive my writing as particularly authentic. I believed I was merely recording a chronological sequence, pruning the branches of memory for clarity. Undoubtedly, I genuinely appreciated my grandmother’s house during that period—though now, describing it in such terms might seem somewhat juvenile....

October 7, 2024 · 3 min · A1yCE

Faraway_ch

初中习作时,写了一篇描写住在乡下的外婆的文章,语文老师看了很受感动。然而,那时我并不觉得那文章有多真诚。我觉得自己只是流水账式地摹写故事,修...

October 4, 2024 · A1yCE

Some Reflection on the Iterating Method of Large Model: Parameter Count or Multimodality

Intro Nowadays, people are familiarized with large model, such as LLMs, in their daily lives. Discussions about which model should own the title of “strongest” are prevalent. AI startup enterprises are striving to optimize their models by any means necessary, including scaling model’s parameter count. As diving into this field of deep learning for more than one year, I have grown weary of this competition focused on model’s parameter count (Hereinafter referred to as NoP, Numbers of Parameter)....

September 23, 2024 · 3 min · A1yCE
test alt content

Some Foundemental Tests

September 22, 2024 · 1 min

Legacy.01

晚自习时下起了雨,我想与的同桌换个位置。我斟酌了很久,也许我该说,让我看看这雨。 但我并没有说。 我还是离开教室,漫无目的地走着,在大雨里本身就...

March 26, 2024 · 2 min · A1yCE.L

Legacy.02

心悸与白雏菊 —— 宽容不了过去,理解不了未来。 想要离开,往昔的枝叶无处不在。我们的城市,是巨树上支离破碎的虚荣。 她走入危楼的影子,城市之心开始...

December 30, 2023 · 3 min · A1yCE.L

Legacy.03

提线木偶 小到轻轻作响的怀表,大到远在北欧的强子对撞机,你看到这些极度精密的仪器时,会不会有想知道它们内部运行原理的冲动?我不知道这是我的性格...

August 31, 2023 · 3 min · A1yCE.L

Legacy.04

散漫言辞·再续·其一 2023.8.26 01 故事大多如此,我们也很难见到畅快淋漓的结局。 02 风车之花 03 我们支离破碎的记忆,理当与风同葬。 04 Everything,how it ends,just like how it starts. 状若终末,...

August 26, 2023 · 1 min · A1yCE.L