Resource directory
Find what you need for your workflow.
Guides: techniques, setup, and troubleshootingSearch Sweet Tea Pipes, external workflows, models, node packs, and guides in one evidence-linked catalog.
Start with a workflow instead ↗Search results
All resources
15 shown of 31,311
Gemma 4 26B A4B It DASHQ INT4 G64
> DASH-Q — Diagonal-Aware Shrinkage for Robust PTQ. > INT4 · group size 64 · 16.4048 GB (from 51.6120 GB — 3.1x smaller)
Gemma 4 26B A4B It DASHQ INT4 G32
> DASH-Q — Diagonal-Aware Shrinkage for Robust PTQ. > INT4 · group size 32 · 17.9357 GB (from 51.6120 GB — 2.9x smaller)
Gemma 4 31B It DASHQ INT2 G32
> DASH-Q — Diagonal-Aware Shrinkage for Robust PTQ. > INT2 · group size 32 · 14.9555 GB (from 62.5463 GB — 4.2x smaller)
Gemma 4 31B Redteam
This model is superseded by dazeb2/Gemma-4-12B-redteam-v5.
Minimax M2 Expert Prune Heal
A research artifact on structurally compressing a 230B Mixture-of-Experts coding agent (MiniMax-M2: 256 experts/layer, top-8) by pruning experts — keeping it a working agent in a real tool-using loop, not just a...
Qwen3.6 35B A3B DASHQ INT4 G64
> DASH-Q — Diagonal-Aware Shrinkage for Robust PTQ. > INT4 · group size 64 · 21.8928 GB (from 71.9038 GB — 3.3x smaller)
Diffusion Aloha Transfer Cube Seed42
Trained with LeRobot. Date: 2026-05-14 12:51 Policy type: diffusion | Device: cuda
Slm 125m Base
A 125M-parameter Llama-style base language model, pretrained from scratch on a legal/financial corpus (US case law + SEC filings + a web slice, 2.19B tokens).
Zerocloud Models
This repository hosts quantized models for my application.
Krea 2 Base Turbo NVFP4 FP8 INT8
Quantized versions of Krea 2 Base and Krea 2 Turbo, converted using silveroxides/converttoquant (ctq CLI) with ComfyUI-compatible metadata.
Nemotron 3.5 Asr Streaming 0.6b Onnx
openmdw-1.1
Gemma 4 31B It DASHQ INT3 G32
> DASH-Q — Diagonal-Aware Shrinkage for Robust PTQ. > INT3 · group size 32 · 19.3485 GB (from 62.5463 GB — 3.2x smaller)
GLACIER 100k MiniMol
GLACIER is a multimodal student-teacher foundation model designed for molecular property prediction. It integrates molecular graphs, SMILES strings, and physicochemical descriptors to learn rich molecular embeddings.
Flowstate R1 GGUF
Pure Rust converter and inference engine for ibm-granite/granite-timeseries-flowstate-r1.
Gemma 4 26B A4B It 8bit
8-bit MLX quantization of google/gemma-4-26B-A4B-it, for Apple Silicon (26 GB). Sparse-MoE vision-language model — run it with mlx-vlm, not mlx-lm. The expert router (router.proj) is kept at 8-bit precision.
