AirLLM 70B inference with single 4GB GPU

Run OpenClaw AI agent with zero API cost, local LLM via AirLLM

Run 70B+ LLMs on a single 4GB GPU — no quantization required.

Moved to here: https://github.com/lyogavin/airllm

No description available

Run 70B+ LLMs on consumer GPUs using layer streaming

AirLLM++: Faster than AirLLM with super-block residency and deep prefetching

🚀 Optimize memory for large language models, enabling 70B models on a 4GB GPU and 405B Llama3.1 on 8GB VRAM without compression techniques.

this is my first personal project

AirLLM 70B inference with single 4GB GPU

A TUI for AirLLM

No description available

OpenAI compatible server for AirLLM

High-performance layer-wise LLM inference in Rust. Memory-efficient inference for models that don't fit in GPU memory.

OpenAI-compatible API server for running Qwen3.5-397B-A17B locally using AirLLM.

Quick test of AirLLM - running 405B parameter model on a laptop

AIRLLM support for amd graphics cards

Run GLM-4.7 (358B) on 8GB VRAM with AirLLM

Rust Layer-Streamed Inference Runtime (AirLLM-style control plane)

No description available

airllm test

No description available

No description available

No description available

AirLLM Docker container

Local-first AI debugging assistant for hardware verification (Buscraft++)

this is a sub repo of the main project

No description available

No description available

No description available