Cadenya
Freemium
Developers building agentic workflows
AutoGPU
FreeOpen SourceSelf-HostedAgentic
Chip design projects stall when EDA toolchains require a specialist for every loop iteration — timing closure, RTL revisions, physical layout tweaks — turning what should be an experiment into a six-week calendar block. AutoGPU is an open-source agentic framework that puts that iteration loop inside an LLM-orchestrated agent swarm.
Similar tools
CadenyaFreemiumCadenya is an agent runtime built around a layered model: tools connect once through MCP…
WorldsFreemiumWorlds spins up a sealed twin of the external systems your agent touches — Stripe and Zendesk…
Brain MemoryFreeBrain Memory stores agent decisions as Markdown files with YAML frontmatter, organized in a…
JevProxyFreemiumThe proxy sits in front of your Cursor or Claude Code agent fleet and classifies incoming tool…
Get This Tool
License: License: unverified
Local-run terms: Source code and Markdown agent instructions publicly available on GitHub; users can clone and run locally with appropriate EDA tools.
Screenshots 2
Summary
The repo describes autonomous agents writing RTL, running it through real EDA tools, reading timing and layout reports, and revising the design — iterating without a human in the seat for each pass. The documented target is small systolic array architectures, specifically matrix-multiply accelerators; the codebase includes ISA definitions, physical design configs, and golden reference models. At that constrained scope, researchers report the agent loop closes. Scale the design complexity beyond what the existing module hierarchy covers and the agents lose the plot — the feedback loops that work for a mac array do not generalize to a multi-block SoC. Teams pushing past the documented scope end up writing their own agent scaffolding on top, at which point AutoGPU is a reference rather than a runtime.
Bottom line: Pick this when you need an agentic baseline for LLM-driven RTL-to-layout experiments on small array designs; plan a different architecture the moment your target design spans more than a handful of modules or requires sign-off flows the existing agent loop does not encode.
Find your fit
Before you choose AutoGPU
More options in Agent Frameworks. Compare pricing, platforms, and features side by side.
Brain Memory
Free · Open source
Long-term continuity in multi-session coding projects
Pros
- Full-stack agentic loop from RTL generation through physical layout hardening, so you avoid the manual handoff between code generation and EDA execution that makes most LLM hardware tools a partial solution.
- Ships with ISA definitions, module RTL, and golden reference models for matrix-multiply accelerators, which means the agent has structured domain context on day one rather than hallucinating architecture details from scratch.
- Entirely open-source with no paid-only features, so the full agent scaffolding, EDA integration hooks, and design configs are auditable and forkable — no black-box inference calls gating the loop.
- Self-hosted by default, which means your RTL, timing reports, and design IP stay on your own infrastructure rather than transiting a vendor's API.
- Iterative revision loop reads real EDA output — timing reports, layout feedback — and feeds it back into the agent, so design errors surface and get corrected inside the automated loop rather than piling up for a human review session.
Cons
- The agent's planning and feedback parsing are scoped to the existing module hierarchy — small systolic arrays and mac structures. When a design introduces module types outside that vocabulary, the agent loses coherent planning context and the loop stalls or produces nonsense RTL; teams at that point are extending the framework from source, not using it.
- No API surface and no abstraction layer between the agent and the raw EDA toolchain means EDA tool version changes or environment differences break the agent loop silently; debugging requires tracing through agent execution logs and EDA stdout, not a structured error interface.
- Star and fork counts from the repository indicate this is an early-stage research artifact with a single primary contributor — community-reported workarounds, tested configurations, and maintained documentation are sparse, so teams that hit an undocumented edge case have the source code and nothing else. Teams needing a maintained, production-grade EDA automation layer with active support will move to a commercial EDA vendor's scripting environment instead.
About
- API Available
- No
- Self-Hosted
- Yes
- Last Updated
- June 11, 2026
Best For
Who it's for
- Researchers exploring agentic chip design
- Teams experimenting with LLM-orchestrated EDA flows
- Hardware prototyping of small systolic arrays
What it does well
- Autonomous hardware accelerator design
- Agent-driven RTL generation and verification
- Iterative physical design optimization with real EDA tools
Frequently Asked Questions
- Is AutoGPU free?
- AutoGPU offers a free option. Hosting, model usage, or optional services may have separate costs; check the pricing details and official terms.
- Is AutoGPU open source?
- Yes. AutoGPU is open source.
- Can I self-host AutoGPU?
- Yes. AutoGPU supports self-hosting on your own infrastructure.
- When was AutoGPU released?
- AutoGPU was first released in 2026.
Curated lists that include this category
AutoGPU wires LLM agents into a closed loop with real EDA tooling: agents generate RTL, invoke physical design steps, parse the resulting timing and layout reports, and resubmit revised designs — repeating until constraints are met or the loop fails. The repository ships ISA definitions, module-level RTL for systolic array components, Python modeling infrastructure, and experiment scaffolding, so the agent has structured context to reason against rather than a blank slate. The workflow is agentic in the strict sense: planning, execution, and revision happen without a human signing off each step.
The differentiating claim, stated directly in the README, is that the agent swarm handles the full stack from RTL through 3D layout hardening — not just code generation that stops before synthesis. Most LLM-assisted hardware tools halt at RTL output and hand off to a human for physical implementation. AutoGPU’s agent loop is designed to traverse that boundary, which is the specific problem it was built to attack.
The practical scope is narrow by design. The documented architecture targets small matrix-multiply arrays and systolic grid structures; the module hierarchy in the repository reflects that. Researchers exploring agentic chip design at that scale have a working baseline with real EDA integration. Teams expecting the agent to generalize to larger or structurally different designs will find that the agent’s planning context and feedback parsing are tuned to the existing module vocabulary — novel architectures require extending that context, which means contributing to the framework rather than configuring it. There is no API, no hosted version, and no graphical interface; this is a self-hosted, CLI-driven research tool.
Contribute
- Work on AutoGPU? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.