Agents, the harness that runs them, and the inference stack underneath.
We build a TypeScript agent library, a coding harness on top of it, and we do original systems research to make open models run faster on real hardware.
Adaptive speculation: draft length that follows measured acceptance
Standard speculative decoding drafts a fixed number of tokens and hopes the model agrees with them. On Qwen3.8-27B (a new open-weight model roughly on par with Claude Opus 4.6), a fixed draft length collapses under low acceptance. Our --spec-draft-adaptive patch to llama.cpp scales draft length to the model's actual acceptance rate instead, live.
3.2× faster than fixed-length speculation at the same draft budget. 4.7× the bare, non-speculative decode rate.
One foundation, four things that run on it.
Agention
A TypeScript-native agent library. Claude, OpenAI, Gemini, Mistral, and local models via Ollama and llama.cpp behind one interface: composable, observable, no hidden state machines.
Read the docs ↗Marshall
A coding harness built on Agention. Name an agent, pin a model, give it a job: planner, coder, reviewer, each running whatever model suits it, local or hosted.
Try Marshall ↗Adaptive Speculation
A fork of llama.cpp with adaptive-length speculative decoding and a Vulkan backend tuned for AMD's unified-memory APUs. Upstream contribution in progress.
Read the research →Quantized Models
Requants and Strix Halo-specific builds, published on HuggingFace: one mainline-compatible, one using our ROCmFPx format for maximum quality-per-byte on unified memory.
See the models →Bricks: a floating city, shipped end to end by the harness.
Bricks is a brick island suspended in the clouds, under zombie siege. It was built using Qwen3.8-27B running through Marshall: score, health, ammo, orbit camera, day/night cycle, the works. It's a lab experiment, a small, playable proof of what the harness can ship on its own.
Play Bricks ↗-
March 22, 2026
Multi-Provider AI: When and Why
A practical guide to using multiple AI providers and when it makes sense for your application.
-
January 15, 2026
Getting Started with Agention
Learn how to set up your first AI agent with Agention and start building multi-provider applications.
-
July 8, 2025
Building Type-Safe Agent Pipelines
How to use TypeScript generics to build robust agent pipelines with full type inference.