--- type: "post" id: "compass-is-coming-to-the-cloud-analysis-bea975" slug: "compass-is-coming-to-the-cloud-analysis-bea975" title: "Compass is coming to the cloud" description: "Compass Cloud shifts Cohere’s retrieval stack from a primarily self-hosted and internally deployed capability into a managed integration surface for enterprise RAG and agent workflows." retrieval_nugget: "Compass Cloud shifts Cohere’s retrieval stack from a primarily self-hosted and internally deployed capability into a managed integration surface for enterprise RAG and agent workflows." published_at: "2026-09-25" updated_at: "2026-09-25" record_date: "2026-09-25" date_kind: "published_at" topics: [] source_urls: ["https://cohere.com/blog/compass-cloud-beta"] schema_version: "newruntime-agent-readable-v0.2" stable_id: "post:compass-is-coming-to-the-cloud-analysis-bea975" status: "published" visuals: [] editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:e838114a6fd0a9e6a3678106fdc8680943612cf1cb6f834b32609b24698df559","reviewed_at":"2026-09-25T16:13:54.298Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":2,"related_record_count":0} analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Compass Cloud shifts Cohere’s retrieval stack from a primarily self-hosted and internally deployed capability into a managed integration surface for enterprise RAG and agent workflows.","observed_facts":[{"text":"Cohere announced Compass Cloud as a private beta managed offering with a limited number of enterprise beta partners.","source_urls":["https://cohere.com/blog/compass-cloud-beta"]},{"text":"Compass Cloud is accessible through APIs, an MCP server, and a Python SDK; self-hosted deployments remain available for privacy-constrained projects.","source_urls":["https://cohere.com/blog/compass-cloud-beta"]},{"text":"Cohere reports a 81.1 versus 64.8 nDCG@10 result against Azure Search on its internal High Finance benchmark.","source_urls":["https://cohere.com/blog/compass-cloud-beta"]}],"mechanism":"The mechanism is operational consolidation: Cohere operates the retrieval pipeline and model inference while an application calls a unified service or exposes its retrieval through MCP.","why_now":"That matters now because agentic workflows can issue repeated, multi-hop searches, multiplying both latency and the cost of irrelevant context.","implications":["A managed retrieval layer could lower the integration burden for enterprises building agents while making vendor choice at the retrieval boundary more consequential.","Teams can evaluate Compass as an MCP-accessible retrieval service rather than assembling separate ingestion, indexing, ranking, and permissions layers."],"evidence_boundary":"Cohere reports 81.1 nDCG@10 for Compass versus 64.8 for Azure Search on its internal High Finance benchmark, but this is a vendor-produced comparison with workload-specific methodology rather than independent validation.","watch_conditions":["Watch whether the beta gains broad availability, publishes pricing and service limits, and demonstrates permission enforcement and retrieval quality across external enterprise workloads.","Watch for independently reproducible benchmarks and details on supported connectors, regional availability, and retention controls."],"related_records":[],"new_branch_reason":"This begins a distinct record because Compass Cloud introduces a managed, MCP-accessible deployment model for Cohere’s retrieval platform; no prior New Runtime record was supplied for comparison."} editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]} routes: {"html":"https://newruntime.com/posts/compass-is-coming-to-the-cloud-analysis-bea975/","markdown":"https://newruntime.com/posts/compass-is-coming-to-the-cloud-analysis-bea975.md","json":"https://newruntime.com/posts/compass-is-coming-to-the-cloud-analysis-bea975.json"} --- # Compass is coming to the cloud ## Retrieval answer Compass Cloud shifts Cohere’s retrieval stack from a primarily self-hosted and internally deployed capability into a managed integration surface for enterprise RAG and agent workflows. Compass Cloud shifts Cohere’s retrieval stack from a primarily self-hosted and internally deployed capability into a managed integration surface for enterprise RAG and agent workflows. The supplied announcement says Compass Cloud is in private beta with a limited number of enterprise beta partners. It also says developers can use Compass through APIs, an MCP server, or a Python SDK, and that self-hosted deployments remain available for privacy-constrained work. Cohere describes the platform as combining connectors, parsing, dense and sparse representations, hybrid search, reranking, and retrieval-time access controls. The mechanism is operational consolidation: Cohere operates the retrieval pipeline and model inference while an application calls a unified service or exposes its retrieval through MCP. That matters now because agentic workflows can issue repeated, multi-hop searches, multiplying both latency and the cost of irrelevant context. For teams, managed retrieval can move engineering attention from operating components to evaluating answer quality, permission behavior, and integration fit. A managed retrieval layer could lower the integration burden for enterprises building agents while making vendor choice at the retrieval boundary more consequential. Cohere reports 81.1 nDCG@10 for Compass versus 64.8 for Azure Search on its internal High Finance benchmark, but this is a vendor-produced comparison with workload-specific methodology rather than independent validation. Watch whether the beta gains broad availability, publishes pricing and service limits, and demonstrates permission enforcement and retrieval quality across external enterprise workloads. Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json ## Continue in the graph Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`. - collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis. - collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph. - whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)