The Kaitchup – AI on a Budget

Home
Notes
AI Notebooks
The Kaitchup Pro
The Kaitchup's Book
AI Toolboxes
Tutorials
Models
LLM Leaderboards
Archive
About
2-bit VPTQ: 6.5x Smaller LLMs, >95% Accuracy
Very accurate 2-bit quantization for running 70B LLMs on a 24 GB GPU
 • 
Benjamin Marie
2
The Weekly Kaitchup #76
GRPO/TRL - SmolVLM - Coconut
 • 
Benjamin Marie
DeepSeek-R1: Reinforcement Learning with Simple and Verifiable Rewards
Qwen2.5 and Llama 3.x are good students
 • 
Benjamin Marie
1
Estimating Memory Usage for LLMs During Inference (V2)
KV cache, GQA, FlashAttention, activations, batching...
 • 
Benjamin Marie
6
The Weekly Kaitchup #75
MiniMax-01 - Qwen2.5 PRM - Kokoro and OuteTTS
 • 
Benjamin Marie
MMLU: Do LLMs Really Know?
When 1. = １. = A. = Ａ. = a. = A) disrupts LLM's world knowledge and language understanding capabilities
 • 
Benjamin Marie
3
Local Agentic AI with smolagents and Qwen2.5 Coder
When to use it and when does it fail
 • 
Benjamin Marie
3
The Weekly Kaitchup #74
Open Phi-4 - OLMo 2 Paper
 • 
Benjamin Marie
RTX 50 and DIGITS: What Does It Mean for Local AI?
Fine-tuning a 2-bit Llama 4 70B with a single consumer GPU
 • 
Benjamin Marie
1
DeepSeek-V3: Understanding and Running the Best Open LLM Locally
A huge but efficient MoE
 • 
Benjamin Marie
1
The Weekly Kaitchup #73
2025 - Smolagents - Bamba
 • 
Benjamin Marie
1
Reasoning with QwQ and QvQ on Your Computer
When "preview" becomes meaningful
 • 
Benjamin Marie
© 2025 The Kaitchup
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great culture