EXPLORE THE CATALOG
Browse by what
you have, not by hype.
The same 157 documented setups, cut five ways: the hardware, the model, the runtime, the memory budget and the task. Each link opens the setups that match.
BY HARDWARE
Start from the GPU.
The cards people report in this catalog, most used first.
BY MODEL
Start from the model.
What contributors actually run, most reported first.
BY RUNTIME
Start from the runtime.
The serving stacks behind the setups.
BY CAPACITY
Start from your machine.
Capacity is the reported total for a setup. Multi-GPU and unified-memory layouts can behave differently.
8 GB or lessLaptops, iGPUs and modest cards9–12 GBEntry-level local inference13–16 GBCurrent mid-range cards17–20 GBRX 7900 XT and compact pro cards21–24 GB3090, 4090 and 7900 XTX class25–32 GB5090 and workstation cards33–48 GBWorkstation and multi-GPU setups49–96 GBLarge multi-GPU and datacenter setups97 GB+Unified memory, servers and clusters
BY TASK
Start from the work.
Concrete workflows, including the tradeoffs their authors reported.
START FROM YOUR CASE
Can I run it?
Give a capacity band and a task, and see the published setups within it.