Dlgo

Dlgo is a Go inference engine and deep learning primitive library for running GGUF models locally on CPU or GPU.

Dlgo is a fast LLM inference engine written in pure Go that lets users run GGUF models locally on CPU or GPU without requiring Python, CUDA, or external inference runtimes. It includes support for multimodal vision inference, text-to-image diffusion pipelines, speech-to-text with Whisper, and voice activity detection with Silero VAD. The platform features an interactive chat interface and an HTTP server with an OpenAI-compatible API. It is designed for developers who want to embed inference directly into their Go applications or run models locally with hardware acceleration. Key architectural highlights include Vulkan GPU compute backends, a never-OOM GPU fallback system, and support for mixture of experts and hybrid recurrent layers.

Dlgo at a glance

Best for: Software engineer
Runs on: Web, API, CLI, self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: github.com/computerex/dlgo

Key features

25+ quantization formats including Q40, K-quants, and I-quants
Vulkan GPU inference with quantized MatVec shaders
Never-OOM GPU automatic VRAM budget with fallback
Mixture of Experts fused multi-expert GPU dispatch
Hybrid SSM and Attention recurrent layers
Speech-to-text transcription with Whisper
Voice activity detection with Silero VAD
OpenAI-compatible HTTP server and API

Use it for

Run local chat sessions and interact with GGUF models
Execute local text-to-image diffusion pipelines
Process multimodal vision and image inputs
Perform speech-to-text transcription
Embed AI inference directly into Go applications

Alternatives to Dlgo

AirLLM — Runs massive 70B+ parameter models on small 4GB consumer GPUs.
llm-d — Deploys production-grade LLM serving on existing hardware clusters.
LMDeploy — Accelerates inference with high-performance engines and quantization tools.
Reflex — It is optimized for cold-start latency and single-tenant decisions.
audio.cpp — Runs modern audio models locally with high-performance, native C++ execution.
fal — Access 1,000+ generative models and high-performance serverless GPUs.
WebLLM — Runs high-performance language models directly inside web browsers.
Soup — Fine-tune and align models like Llama-3.1-8B on a 4 GB laptop GPU.
Strata — Runs large AI models locally on consumer gaming PCs without server reliance.
vLLM — Maximizes throughput with PagedAttention and continuous batching.
Resin — Builds high-performance inference engines for complex models like DeepSeek and GLM.
whichllm — Ranks models using real benchmarks and hardware fit instead of parameter count.

All Dlgo alternatives