What is zero data retention (ZDR)?

Zero data retention (ZDR) is the claim that an AI endpoint does not keep the prompt, the completion, or other residue of a request after the call finishes.

Fireworks AI open models zero data retention: terms and exceptions

Q: What is the ZDR score for Fireworks AI open models?

On zdr.dev the published score is ZDR 0.5 (f) · B. ZDR by default, volatile memory only. Responses API defaults store=true -> 30d; proprietary Fireworks models may log.

ZDR 0.5 (f) · B

Short answer

ZDR by default, volatile memory only. Responses API defaults store=true -> 30d; proprietary Fireworks models may log.

Scope and evidence

Provider	Fireworks AI
Service	open models
Lane	ZDR
Origin	US United States
License	open-weight
Hosting	Self-host
Score	ZDR 0.5 (f) · B Architectural ZDR. Nothing durably written by provider; stateful features blocked/rejected; still third-party plaintext processing
Evidence	Public documentation reviewed (B: published documentation)
Source-review date	2026-08-24 (registry)

Other lanes for Fireworks AI

ZDR 0.5 (f) · Bfireworks/api · open modelsZDR

Structured breakdown

Default posture: ZDR by default, volatile memory only
Exceptions and carve-outs: Responses API defaults store=true -> 30d; proprietary Fireworks models may log

Approval and setup path

Approval and setup path not yet reviewed.

Flags on this record

f: stateful-feature carve-outs (files, batch, threads, caching, grounding)

Practical questions for your agreement

Does your signed agreement cover this exact service, model set, and feature path?
Which retention, logging, and human-review exceptions still apply after any ZDR election?
What configuration evidence (flags, project settings, store defaults) confirms the posture you expect?
Do your customer commitments match what the reviewed evidence supports?