tensorfold

NVFP4 BIZ: nvidia/GLM-5.3-Flash-NVFP4 as distributed, on DGX Spark-class GB10 systems. 2.x serves it with TensorFold (TP=2 or TP=3, FP8 KV, MTP, drafted replies equal serial); 1.x with a pinned vLLM (TP=2 or TP=3, images, optional AXL repack). Apache-2.0 code, MIT weights fetched separately. BIZ = business-use intent, not support or certification.

MiniMax H3 audio-video model on TensorFold NVFP4 kernels: a faster ComfyUI loader for one RTX 50-series GPU (2.3x stock at 20 steps, 81 s per 5 s 1344x768 video with Turbo + sparse attention on an RTX 5070 Ti)

GLM-5.3-Flash EXL3 on 4x RTX PRO 6000 with TensorFold: one-command recipe and engine patches. By Aevonix Research in collaboration with Mia's AI Lab.

Qwen3.8-27B AEON Ultimate Uncensored (4-bit MLX) on TensorFold 0.6.0 + DFlash2, Mac Studio M5 Ultra: 191 tok/s single-stream decode, 97 ms TTFT, 1.4k tok/s prefill; one-shot + GHCR carrier

Reproducible MLX Q4 conversion of Cantina Security Apex Flash 1 for TensorFold on dual DGX Spark

One command: GLM-5.3-Flash with the Dealign o_proj abliteration transplant on Mia's TensorFold recipe (2x DGX Spark). Byte-verified, stock-speed decode.

Qwen3.8-27B on two NVIDIA DGX Sparks (TP2) with TensorFold: one setup script, benched (99 tok/s single stream, 387 tok/s at 8 users)

NVIDIA DGX Spark LLM setup — start here: run Qwen and Gemma locally on 1 or 2 DGX Sparks (GB10) with TensorFold; tested recipes, measured speeds, cabling

TensorFold v0.6.0 GLM-5.3-Flash engine generalized from 2 to 4 DGX Sparks (TP4) over a switchless RoCE fiber ring - uncensored EXL3 checkpoint, overlay + tooling. Upstreams: ashhart/TensorFold, MiaAI-Lab 2x-Spark stack, alexellis/switchless-nccl.

Qwen3.6-35B-A3B on one NVIDIA DGX Spark with TensorFold: one setup script, fits 64 GB Sparks, benched (160 tok/s single stream, 559 tok/s at 8 users)