dlgo runs GGUF models locally in pure Go, on CPU or GPU, with no Python, CUDA or external inference runtime required. It installs as a small set of commands: a core chat CLI, a web server with a UI, plus vision and image commands. It can pull a catalog of compatible Unsloth GGUF releases from Hugging Face into a platform cache, or you can point it at any compatible local GGUF file or a direct HTTPS URL. GPU and Vulkan acceleration needs a source build with the vulkan tag and a cgo toolchain.
Be the first to rate dlgo. Your feedback helps other users decide.
Install the CLI and run a local GGUF file, the way you would with a self-hosted runner.
Start the server command and use the bundled web interface against the model.
Download a compatible Unsloth release by alias, or pass any local GGUF path or HTTPS URL.
Similar tools in the same category — compare side-by-side
Ret is a desktop workbench for Codex and Claude Code. You start threads across every project, fork t...
glossdex is private semantic search over a folder of photos and documents that returns nothing when ...
RemainderBot spends the Claude or Codex quota you were about to lose. Your subscription resets every...
Guides and comparisons about code tools
Ten launches that keep the model, the index and the facts on your own machine, from search that admits no match to GPU training in CUDA.
Eight Show HN launches where the agent owns the device rather than the editor: desktop control, phones, memory pressure, kernel security and live repair.
Ten Show HN launches that turn an opaque agent run into something you can see: step progress, session state, approvals, image paste and voice.
Ten launches that put a number on agent work: per-repo bills, live cost bars, expiring quota, cheap worker models and local decision engines.
Nine launches that draw the line around an agent: policy proxies, guard hooks, semantic permissions, sandboxes, worktrees and a proof checker.
