dspark

Up to 4× faster LLM decoding on Apple Silicon, lossless. Native MLX port of DeepSeek's DSpark & z-lab's DFlash speculative decoding — Gemma-4, Qwen3.8, Muse-Glimmer, Nemotron, LFM2.5, Ornith-1.0, ternary Bonsai-27B.

A native .NET LLM inference engine for GGUF models. TensorSharp provides a console application, a web-based chatbot interface, and Ollama/OpenAI-compatible HTTP APIs for programmatic access. It supports Windows/MacOS/Linux with full GPU capability

A PyTorch native library for training speculative decoding models

DeepSeek V4 Flash 284B on AMD Strix Halo (gfx1151) — up to 32 tok/s decode & ~250 tok/s prefill via ROCmFPX, DSpark & ROCm 7.2

Optimized two-node DGX Spark deployment recipe for DeepSeek V4 Flash with vLLM, DSpark, and NVFP4 KV cache

DeepSeek-V4-Flash-0731 on 8x RTX 3090 (SM86) with vLLM — verified FP8 serving, benchmarks, build guide, and reproducible release.

Anemll + MiaAI-Lab dual DGX Spark DSV4F+DSpark on vLLM 0.25.2.dev0 — 100% abliterated weights, Spec-Kit one-shot agent package. Full credit Anemll/MiaAI.

DeepSeek-V4-Flash + DSpark speculative decoding on a pair of NVIDIA DGX Sparks (vLLM TP=2 over RoCE) — tuned recipe, overlays that halve multi-turn TTFT, contamination-guarded benchmarks, ops runbook

MiniMax M3 NVFP4 on 4x DGX Spark with NVIDIA DSpark, native multi-node vLLM TP=4, reasoning, and tool calling

Lossless inference speedup benchmarking suite for local LLMs using DeepSeek DSpark speculative draft heads.

Source-only offline deployment, testing, and operations toolkit for DeepSeek-V4-Flash-0731 on 4x/8x NVIDIA A100 GPUs, pinned to a reviewed community vLLM R1 stack.

Code to train and evaluate DSpark draft models for speculative decoding using the speculators library and vllm

DSpark-style speculative decoding for OCR VLMs on Apple Silicon (MLX). Faster local OCR for DeepSeek-OCR-2, GLM-OCR, Unlimited-OCR.

DeepSeek-V4-Flash-Vision-Exp on 2x Jetson AGX Thor (SM110): Thor-adapted vLLM port, verified + benchmarked. Native image input with DSpark speculative decoding, TP=2 dual-node.

DeepSeek V4 Flash 0731 on one RTX PRO 6000/WSL2: CUDA 13.3.1, vLLM-Moet, native OpenCode tools, DSpark-5 greedy, 128K FP8 KV, plus clean-MAX fallback.

Verified IQ2_XXS plus native DSpark recipe for DeepSeek V4 Vision on one DGX Spark.

Universal Dual-Engine & Speculative Curation Configuration Kit for Antigravity, Cursor, Claude Code, Windsurf & Grok (FastMCP)

Verified two-node GB10 DeepSeek-V4-Flash-0731 DSpark Graph-8 deployment and benchmarks

Agent-Level Speculative Orchestration & Formal Dual-Engine Code Generation (Rust, FastMCP, DeepSeek)

Exactness-first fast iteration lab for improving DSpark inference speed