DwarfStar 4 (ds4) is a narrow C inference engine for running frontier open-weight models locally on high-memory Mac, CUDA and ROCm machines. It supports DeepSeek V4 and V4.1 Flash, GLM 5.x and Qwen3.8 Flash Next, with text and vision models, local APIs, a CLI and a native agent in one stack. MIT licensed.
Be the first to rate DwarfStar 4. Your feedback helps other users decide.
Serve DeepSeek V4 or V4.1 Flash from your own machine
Use text and vision models through the same local API and CLI
Run the bundled native agent entirely on your hardware
Similar tools in the same category — compare side-by-side
MemorySync gives AI agents persistent memory through an API. In the browser playground you tell it o...
Jevcrawl is an MCP server and CLI that crawls websites for AI agents with no LLM in the crawl loop. ...
MCPJam is a testing, debugging and evaluation platform for MCP servers. It covers OAuth flows, LLM-p...
Guides and comparisons about code tools
Ten October 2026 agent infrastructure launches: local inference engines, sandboxes that report their isolation boundary, agent memory and MCP testing.
Discover the best AI code tools for testing automation in 2026. Generate tests, find bugs, and automate QA workflows with AI-powered testing solutions.
Explore the best AI code tools for deployment automation in 2026. From CI/CD pipeline generation to infrastructure as code, streamline deployment.
Inventory, planning, collision detection and scheduled loops: recent open-source tools for running several coding agents against one repository.
Code graphs, persistent decision memory, semantic layers and artifact plumbing: recent tools giving coding agents the context they keep losing.