tensor-parallelism

🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading

InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencies.

Slicing a PyTorch Tensor Into Parallel Shards

Decentralized LLMs fine-tuning and inference with offloading

Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*

An Efficient and Versatile Inference Engine for Distributed LLM Serving

JORA: JAX Tensor-Parallel LoRA Library (ACL 2024)

A distributed training framework for large language models powered by Lightning.

Tensor-parallel DeepSeek V4 Flash inference on dual AMD Strix Halo over OdinLink USB4/TB5 RDMA or Mellanox RoCE v2.

DeepSeek-V4-Flash-0731 tuned on 4x DGX Spark (GB10) at TP=4 — 123.13 tok/s decode. Every variable measured one at a time, with the values that lost.

Tencent Hy3 295B MoE (NVFP4) on 2x NVIDIA DGX Spark — TP2 over 200GbE, 256K context, 26 tok/s end-to-end. Tuned, benchmarked, agent-ready.

Fast and easy distributed model training examples.

Tensor Parallelism with JAX + Shard Map

DeepSeek-V4-Flash DSpark speculative decoding on 2x DGX Spark (GB10/sm_121), 1M context — our fp8 recipe + independent reproduction and cross-build benchmarks of the NVFP4-KV build. Honest, apples-to-apples measurements.

DeepSeek V4 Flash 0731 on 8x NVIDIA DGX Spark: deployment, reproducible benchmarks, optimization profiles, and sanitized evidence.

Running large LLMs on pre-Ampere NVIDIA hardware — Tesla V100 (sm_70), RTX 2080 Ti (sm_75), CMP 170HX. Measured benchmarks, vLLM forks, and the hardware side: NVLink on SXM2 carrier boards, driver traps, cooling, used-kit acceptance.

Multi-GPU tensor-parallel vLLM on AMD Radeon RX 7900 XT / XTX / GRE (7900XT, 7900XTX, RX7900XT, gfx1100, RDNA3, ROCm): root cause and fix for the RCCL hostcall / PCIe atomics (AtomicOps) crash "NCCL error: unhandled cuda error" / "operation cannot be performed in the present state", Proxmox VFIO passthrough; LLM inference benchmarks on 13 machines

Fork of llama.cpp Nvida Volta V100 (tensor parallelism 4 x v100 GPU)

GPU Memory Calculator for LLM Training - Calculate GPU memory requirements for training Large Language Models with support for multiple training engines including PyTorch DDP, DeepSpeed ZeRO, Megatron-LM, and FSDP.

Optimized dual AMD Strix Halo (gfx1151) vLLM MoE inference: TP=2 over USB4, tuned int4 MoE kernel, ~8us interconnect, serialized serving