I lost my friend a few weeks ago. Still hard to believe.
I knew him best for his art. For pushing models to the max. Since 2021 I spent hours chatting, tinkering and learning from him.
I have years of DMs I've been working through slowly. Art that most never saw.
This is one of my favorites. Was so far ahead of what people could do at the time. This image is 4 years old. Before most even fiddled with image gen.
A true legend and a gem. My heart is heavy my friend.
Gansy I miss you❤️❤️❤️
💫 GANSY ETERNAL
🧵1/6
3 weeks ago we lost GANSY (@theGANSmachine), a pioneering generative artist, a core contributor to the Ordinals ecosystem, and an incredibly kind soul.
Today we invite you to create art in his memory.
The GANSY Eternal Collage is open.
Read on below 👇
I lost my friend a few weeks ago. Still hard to believe.
I knew him best for his art. For pushing models to the max. Since 2021 I spent hours chatting, tinkering and learning from him.
I have years of DMs I've been working through slowly. Art that most never saw.
This is one of my favorites. Was so far ahead of what people could do at the time. This image is 4 years old. Before most even fiddled with image gen.
A true legend and a gem. My heart is heavy my friend.
Gansy I miss you❤️❤️❤️
💫 GANSY ETERNAL
🧵1/6
3 weeks ago we lost GANSY (@theGANSmachine), a pioneering generative artist, a core contributor to the Ordinals ecosystem, and an incredibly kind soul.
Today we invite you to create art in his memory.
The GANSY Eternal Collage is open.
Read on below 👇
The biggest money-making opportunity in AI isn’t in tech.
It’s in entertainment.
In the next few years, we’ll see a billion-dollar franchise , the next Game of Thrones, Demon Slayer, or Attack on Titan created by a writer and a team of 10 to 20 people.
Imagine the billions generated by Game of Thrones, but split between 10 people instead of thousands. All the rights and royalties going to a single studio/publishing house.
Seedance is only two or three evolutions away from making this possible for small teams.
You don’t want to build the next Cursor, Harvey, or Salesforce.
You want to build the next Disney, Pixar, Marvel, or MAPPA.
@ZssBecker been working mainly in silence for the past 5+ years on this vision. the tools are powerful. almost where we need them. they lack taste though. that is the differentiator.
you need taste, tools and the right people.
@vivaangandhi_ higgsfield is for convenience. nothing is free though. check the ToS and PP.
these days it is easy enough to build custom solutions using apis. fal is perfect for this.
@KingBootoshi @vlelyavin run a system in cloudflare
use if for everything at this point. been fine tuning it for a few months. system runs, i check in for approvals. not bold enough to remove the human review yet😂
The Top 26 Essential Papers (+5 Bonus Resources)
for Mastering LLMs and Transformers
This list bridges the Transformer foundations
with the reasoning, MoE, and agentic shift
Recommended Reading Order
1. Attention Is All You Need (Vaswani et al., 2017)
> The original Transformer paper. Covers self-attention,
> multi-head attention, and the encoder-decoder structure
> (even though most modern LLMs are decoder-only.)
2. The Illustrated Transformer (Jay Alammar, 2018)
> Great intuition builder for understanding
> attention and tensor flow before diving into implementations
3. BERT: Pre-training of Deep Bidirectional Transformers (Devlin et al., 2018)
> Encoder-side fundamentals, masked language modeling,
> and representation learning that still shape modern architectures
4. Language Models are Few-Shot Learners (GPT-3) (Brown et al., 2020)
> Established in-context learning as a real
> capability and shifted how prompting is understood
5. Scaling Laws for Neural Language Models (Kaplan et al., 2020)
> First clean empirical scaling framework for parameters, data, and compute
> Read alongside Chinchilla to understand why most models were undertrained
6. Training Compute-Optimal Large Language Models (Chinchilla) (Hoffmann et al., 2022)
> Demonstrated that token count matters more than
> parameter count for a fixed compute budget
7. LLaMA: Open and Efficient Foundation Language Models (Touvron et al., 2023)
> The paper that triggered the open-weight era
> Introduced architectural defaults like RMSNorm, SwiGLU
> and RoPE as standard practice
8. RoFormer: Rotary Position Embedding (Su et al., 2021)
> Positional encoding that became the modern default for long-context LLMs
9. FlashAttention (Dao et al., 2022)
> Memory-efficient attention that enabled long context windows
> and high-throughput inference by optimizing GPU memory access.
10. Retrieval-Augmented Generation (RAG) (Lewis et al., 2020)
> Combines parametric models with external knowledge sources
> Foundational for grounded and enterprise systems
11. Training Language Models to Follow Instructions with Human Feedback (InstructGPT) (Ouyang et al., 2022)
> The modern post-training and alignment blueprint
> that instruction-tuned models follow
12. Direct Preference Optimization (DPO) (Rafailov et al., 2023)
> A simpler and more stable alternative to PPO-based RLHF
> Preference alignment via the loss function
13. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Wei et al., 2022)
> Demonstrated that reasoning can be elicited through prompting
> alone and laid the groundwork for later reasoning-focused training
14. ReAct: Reasoning and Acting (Yao et al., 2022 / ICLR 2023)
> The foundation of agentic systems
> Combines reasoning traces with tool use and environment interaction
15. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (Guo et al., 2025)
> The R1 paper. Proved that large-scale reinforcement learning without
> supervised data can induce self-verification and structured reasoning behavior
16. Qwen3 Technical Report (Yang et al., 2025)
> A modern architecture lightweight overview
> Introduced unified MoE with Thinking Mode and Non-Thinking
> Mode to dynamically trade off cost and reasoning depth
17. Outrageously Large Neural Networks: Sparsely-Gated Mixture of Experts (Shazeer et al., 2017)
> The modern MoE ignition point
> Conditional computation at scale
18. Switch Transformers (Fedus et al., 2021)
> Simplified MoE routing using single-expert activation
> Key to stabilizing trillion-parameter training
19. Mixtral of Experts (Mistral AI, 2024)
> Open-weight MoE that proved sparse models can match dense quality
> while running at small-model inference cost
20. Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints (Komatsuzaki et al., 2022 / ICLR 2023)
> Practical technique for converting dense checkpoints into MoE models
> Critical for compute reuse and iterative scaling
21. The Platonic Representation Hypothesis (Huh et al., 2024)
> Evidence that scaled models converge toward shared
> internal representations across modalities
22. Textbooks Are All You Need (Gunasekar et al., 2023)
> Demonstrated that high-quality synthetic data allows
> small models to outperform much larger ones
23. Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet (Templeton et al., 2024)
> The biggest leap in mechanistic interpretability
> Decomposes neural networks into millions of interpretable features
24. PaLM: Scaling Language Modeling with Pathways (Chowdhery et al., 2022)
> A masterclass in large-scale training
> orchestration across thousands of accelerators
25. GLaM: Generalist Language Model (Du et al., 2022)
> Validated MoE scaling economics with massive
> total parameters but small active parameter counts
26. The Smol Training Playbook (Hugging Face, 2025)
> Practical end-to-end handbook for efficiently training language models
Bonus Material
> T5: Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (Raffel et al., 2019)
> Toolformer (Schick et al., 2023)
> GShard (Lepikhin et al., 2020)
> Adaptive Mixtures of Local Experts (Jacobs et al., 1991)
> Hierarchical Mixtures of Experts (Jordan and Jacobs, 1994)
If you deeply understand these fundamentals; Transformer core, scaling laws, FlashAttention, instruction tuning, R1-style reasoning, and MoE upcycling, you already understand LLMs better than most
Time to lock-in, good luck ;)
@ChainRacingClub@moltbook i would assume all of them
nothing really new there. been running a similar setup for nearly 5 months. more robust with deeper mem retrieval.
packaged with marketing for the normies so think they're cutting edge.
fun experiment, but there is more interesting things to work on
@AntoineRSX@Zerodartz i leveled up the mac studio and tossed in a mini late last year
nobody else i know is buying one
started picking up spare parts i might need as well. ram, gpu, etc