Open Source · Edge Deployment · Geek-First

Quantize 
Everything.

Bridge the gap between research papers and real-world deployment. Run state-of-the-art LLMs on consumer hardware.

Explore ModelsVRAM Calculator
GGUF
AWQ
EXL2
GPTQ
HQQ
& more
10Models Indexed
5Formats Tracked
33GPUs in Database
99.2%Avg Accuracy Retained

Today's Quant Feed

Latest community quantization releases

View All →
  • NEW
    Qwen2.5-72B-InstructGGUF
    Q4_K_M · 43.6 GB · bartowski
    RTX 4090 ×2·2h ago
  • HOT
    DeepSeek-R1-Distill-Qwen-14BEXL2
    4.65bpw · 9.8 GB · turboderp
    RTX 4090·5h ago
  • NEW
    Llama-3.3-70B-InstructGGUF
    Q5_K_M · 50.1 GB · unsloth
    A100 80G·8h ago
  • UPD
    Mistral-Small-24B-InstructAWQ
    INT4 · 14.2 GB · city96
    RTX 3090·12h ago
  • HOT
    Qwen2.5-Coder-32B-InstructGGUF
    Q4_K_M · 22.0 GB · bartowski
    RTX 4090·18h ago

Format Heat Index

Community adoption · this week

  • 1GGUF
    89%+3%
  • 2AWQ
    45%+7%
  • 3EXL2
    32%0%
  • 4GPTQ
    28%-2%
  • 5HQQ
    18%+12%

vs last week

Quick Tools

VRAM Calculator

Precise memory requirements for any model × quant × context combination. Red/yellow/green hardware verdict.

Open Tool →

CLI Generator

Generate ready-to-run llama.cpp, Ollama, vLLM commands. One-liner or Docker Compose.

Open Tool →

Format Intelligence Radar

Compare quantization formats across 6 key dimensions