{"domain":"doubleword.ai","status":"valid","tier":"listed","authority":51,"verified":false,"title":"Doubleword","summary":"Doubleword is a London-based AI inference lab and provider serving open-weight models through one OpenAI-compatible API at high throughput and low cost, across three tiers: Realtime, Async and Batch. Doubleword's inference systems research - custom kernels, purpose-built speculative decoders, scheduling, quantisation and KV-cache work - is what makes the API cheaper per token than closed-model APIs at comparable quality.","country":null,"country_source":"cdn","llms_txt":"https://doubleword.ai/llms.txt","llms_full_txt":null,"last_checked":"2026-10-19T23:06:32.324Z","agent_terms":null,"categories":["ai-ml","apis-sdks"],"capabilities":[],"inferred_tags":["llm inference","openai-compatible api","open-weight models","batch inference","kv cache"],"audience":"developers","agent_uses":["read_docs","call_api","contact_support"],"language":"en","platform_generated":false,"agent_summary":"London AI inference provider serving open-weight models via one OpenAI-compatible API in Realtime, Async and Batch tiers; full developer docs and CLI guides.","llms_txt_cached":"https://agentwebmap.com/v1/domain/doubleword.ai/llms.txt","agentic":"interface","section_keys":["account-billing","archive","cli","code","conceptual-guides","control-layer","core-workflows","data-sources","docs","doubleword-cli","doubleword-inference-api","file-tools","getting-started","how-to-guides","integrations","models","optional","organizations","overview","projects","reference","research","sdks-cli","site","support","using-the-platform","workbooks"],"section_partial":true,"section_indexed_at":"2026-10-22T20:00:32.835Z"}