Blackwell GitHub Repositories

A high-throughput and memory-efficient inference and serving engine for LLMs

SGLang is a high-performance serving framework for large language models and multimodal models.

TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.

OpenLake is a high performance storage engine for efficient LLM inference and GPU Training

TokenSpeed is a speed-of-light LLM inference engine.

Parallax is a distributed model serving framework that lets you build your own AI cluster anywhere

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

Fully uncensored, capability-enhanced abliteration of Qwen3.6-27B. NVFP4 + z-lab DFlash speculative decoding (n=12) on the unified ghcr.io/aeon-7/aeon-vllm-ultimate:latest container, tuned for long-context draft acceptance on DGX Spark. 6 HF variants (BF16/NVFP4/MTP/MTP-XS), docker-compose, and QuickStart.

A Python DSL to write Nvidia PTX for Hopper and Blackwell in JAX and PyTorch

Rust + CUDA inference engine for NVIDIA RTX PRO 6000 Blackwell and RTX 5090. Serves safetensors and GGUF over an OpenAI-compatible API, with per-device tuned defaults and speculative decode gated byte-identical to plain decode. Hosted instance: inference.tiyuvta.ai

QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning

GLM-5.2-NVFP4-REAP-469B serving on SM120 (4× RTX PRO 6000 Blackwell) — one-command vLLM launch recipe, 250K context, DeepSeek Sparse Attention + MTP speculative decode

Practical local LLM recipes and benchmarks for RTX 5060 Ti setups

One-command vLLM installation for NVIDIA DGX Spark with Blackwell GB10 GPUs (sm_121 architecture)

NVIDIA Sol-Attn for ComfyUI / Triton kernel on SM89 - SM121, with zero-copy MiniMax H3 nodes: memory-efficient attention, scheduled tau with graph preview, and feed-forward chunking. Measured 1.14–1.44× vs SageAttention and −37% MLP peak VRAM on H3

Pre-built wheels for llama-cpp-python across platforms and CUDA versions

Bleeding-edge ComfyUI for NVIDIA DGX Spark (GB10/Blackwell/sm_121a). CUDA 13 + SageAttention v3 (sm_121a) + NVFP4 + 14 custom-node packs + Flux 2 Dev / LTX 2.3 22B / ACE-Step v1.5 XL Turbo pre-bundled with abliterated text-encoder paths.

DFlash vLLM for DGX Spark — Plug & Play Block-Diffusion Speculative Decoding

Prebuilt DeepSpeed wheels for Windows with NVIDIA GPU support. Supports GTX 10 - RTX 50 series. Compiled with pytorch 2.7, 2.8 and cuda 12.8

GLM-5.2 (744B/40B MoE) on a 4× DGX Spark / GB10 (sm_121) cluster: portable Triton sparse-MLA kernels, a data-free expert prune, MTP draft, and a one-script bootstrap.

From-scratch C++23/CUDA inference engine for the NVIDIA RTX 5090 (sm_120a). The best single-GPU backend for agentic AI: tool calling, long-context loops, reasoning and concurrent sub-agents. Decode beats llama.cpp b9976 by 42-48% on dense GGUF (measured 2026-07-12), at-or-ahead of vLLM on NVFP4. 100% written by Claude Code.

A curated list of tools, guides, playbooks, and resources for the NVIDIA DGX Spark (GB10 Grace Blackwell personal AI supercomputer).

No repository description provided.

llama.cpp fork optimized for NVIDIA DGX Spark / GB10 (Blackwell, SM 12.1) — TurboQuant weights + KV, NVFP4, DFlash MTP

GPU-accelerated WhisperX on NVIDIA Blackwell (SM_121) - DGX Spark compatible

Cross-platform FlashAttention-2 Triton implementation for Turing+ GPUs with custom configuration mode

[ACL 2026 Main] Code for the paper "ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs"

Lynn 原生 LLM 推理引擎 · W4A8/NVFP4 量化 · 自写 CUDA/Triton kernel · MoE · 投机解码 | Lynn-native LLM inference engine for NVIDIA Blackwell

DGX Spark research and tests - containers, benchmarks, and investigation notes for running models on GB10 (SM 12.1)

NVFP4 inference on Blackwell GeForce (RTX 5090/5080/5070 Ti/RTX PRO 6000) — SM120 patches for vLLM + FlashInfer + CUTLASS. 175 tok/s on Qwen3.6-35B MoE.