exl3

Open repos and models their makers tagged exl3, most starred first. The same for everyone.

AI on my own machineRepo

kyojin

Kyojin: open-source engine that runs 100 GB-class open models (GLM-5.3, Qwen3.8, DeepSeek V4.1) on one desktop machine. AMD Strix Halo, NVIDIA RTX and Apple Mac. OpenAI API, built on ExLlamaV3.

  • 244
  • Python
  • MIT
  • 2 d ago
Yamz-LabsTake it
AI on my own machineRepo

crucible-strix-halo

Run Qwen3.8-Flash-Next on one AMD Strix Halo (Ryzen AI Max+ 395, gfx1151). EXL3 4.05 and 3.05 bpw packs, own ROCm/HIP kernels, OpenAI-compatible server, 4 sessions of 262K context, setup guide included.

  • 35
  • Python
  • Apache-2.0
  • 8 h ago
jaedyncTake it
AI on my own machineRepo

strix-halo-llm

DeepSeek-V4.1-Flash (1M context) and GLM-5.3-Flash on a BOSGAME M5 (Ryzen AI Max+ 395, Radeon 8060S, 128 GiB) with two RX 7900 XTX over OCuLink: pinned Ansible recipes and measured results. Also the M5-only Qwen and DeepSeek setups.

  • 34
  • Python
  • MIT
  • 8 h ago
pepusczTake it
AI on my own machineRepo

deepseek-v41-flash-exl3-tp3-tp4-spark

DeepSeek 1017 EXL3 on 3-4 DGX Sparks. TP4 4.0.4: 4,277 tok/s cold prefill; C16 code 331.7 / prose 287.5 native decode sum. Smart CUDA Graph, KV Cache and Adaptive Speculation controllers. C16/R48, 9M KV, 3.5 GiB checkpoints. TP3 1.2.8: up to +58% cold prefill. Exact recipes, source and benchmarks.

  • 20
  • Python
  • 2 wk ago
LakyFxTake it
AI on my own machineRepo

GLM-5.3-Flash-Uncensored-EXL3-2x-DGX-Sparks-TensorFold-TP2

GLM-5.3-Flash on two DGX Sparks with TensorFold v0.6.7: the root-caused, bit-exact HC row-split fix, RoCE all-gathers for decode, per-step latency instrumentation, and a working 2-rank NCCL config over a direct CX7 link. Upstreams: ashhart/TensorFold, MiaAI-Lab 2x-Spark stack, alexellis/switchless-nccl.

  • 0
  • Python
  • Apache-2.0
  • 5 wk ago
KeenniTake it
Search GitHub and Hugging Face for exl3

Get , open it once and sign in. Then come back here and press the button again.