VRAM & GPU Cost Calculator
VRAM any HF model needs + live GPU rental prices
🧠Best AI LLMs & AI Infrastructure tools
Models, inference, fine-tuning, evals, vector DBs and MLOps.
Mini-AGI
Dynamic continual learning model trained on 8GB VRAM
Reef
Infrastructure for continually self‑improving agents
JevBench
a reproducible benchmark for typed decision models
Rakazo
Open-source Grok Bot alternative. Choose your own model and sandbox.
Utopia
World's first open-source enterprise world model.
The ultimate guide to multi-harness RL
Train open models with RL inside real agent harnesses
Codex chatgpt web
Use ChatGPT Web (including Pro) as a native model in Codex — with context, tools, streaming and images, without using…
Turbo fieldfare
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
Needle2
14MB agentic LLM for phones, wearables, smart home and robots
ThoughtDAG
An editable context graph for LLM conversations
MiMo-Code
MiMo Code: Where Models and Agents Co-Evolve
Cactus Hybrid
We taught Gemma 4 to know when it's wrong
Yap
OSS on-device voice dictation for macOS with no model to download
Opensquilla
OpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence density
AiSOC
Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable…
Lathe
Use LLMs to learn a new domain, not skip past it
Forge
Guardrails take an 8B model from 53% to 99% on agentic tasks
Needle
We Distilled Gemini Tool Calling into a 26M Model
Semble
Code search for agents that uses 98% fewer tokens than grep
1-Bit Bonsai
the First Commercially Viable 1-Bit LLMs
Now I Get It
Translate scientific papers into interactive webpages
Sweep
Open-weights 1.5B model for next-edit autocomplete
Z80-μLM
a 'Conversational AI' That Fits in 40KB
HN Wrapped 2025
an LLM reviews your year on HN
The Smol Training Playbook
The secrets to building world-class LLMs
Wan2.2 14B Fast
generate a video from an image with a text prompt
The Ultra-Scale Playbook
The ultimate guide to training LLM on large GPU Clusters