Doubleword · Open-weight inference sold by latency tier, plus an open-source gateway teams self-host — the layer where a buy-versus-rent decision turns into a per-token bill.

From readers

How reader reviews work

Ideas and changes

Ask us to test

Name the task. The most backed one goes into the next film.

Hiring

2 openHow Doubleword hires

Member of Technical Staff: Inference SystemsCore Product · London, in person 2 to 3 days per week at our Farringdon office17 days seen
Member of Technical Staff: Inference ResearchInference Lab · London, in person 2 to 3 days per week at our Farringdon office17 days seen

Run Doubleword? Claim this page and reply to readers.

Claim this page

More about Doubleword

Sells inference on open-weight models as a hosted, OpenAI-compatible API at three latency tiers — realtime, async, and 24-hour batch — priced per token, and separately ships an open-source gateway (Control Layer) that teams run themselves.

Sector: Inference
On the roster since: 15 Sept 2026