Bases that are actually yours to keep.

Every base is open-weight. Train it, export it, run it on your own hardware. No licence that expires when the invoice does, and no clause that stops you serving the result.

The catalogue
BaseContextTune time (24 GPUs) FromExport weightsBest at
Penstock 8B128K~4 h$180 YesRetrieval, classification, structured extraction
Penstock 34B256K~14 h$740 YesReasoning-heavy work and long documents
Diffusion XL~6 h$310 YesProduct and brand imagery from a house style
Penstock 1.5B32K~40 min$38 YesEdge and on-device, where latency beats depth
Bring your ownvariesvariesmetered YesAny open-weight checkpoint that loads in transformers

Choosing between them

Start at 8B

Most tasks that read like "understand our documents and answer questions about them" are solved by the 8B. It is cheap enough to run twice and compare, which is worth more than getting the size right first time.

Move up when the eval says so

Go to 34B when the held-out evaluation shows the 8B failing on reasoning rather than on recall. Longer documents alone are not a reason — the 8B already carries 128K.

Bring your own if you have one

If it loads in transformers, the scheduler will pick it up. You keep whatever licence came with it; we do not add one.

Read the training reference