Your Data in GroqCloud - GroqDocs (https://console.groq.com/docs/your-data) citeturn148view2 [wordlim: 200] Crawled: today; Content type: text/html; Source: open({"ref_id":"https://console.groq.com/docs/your-data","lineno":null}); Total lines: 223 L0: 【0†Docs】【1†Login】 L1: L2: ## Documentation L3: L4: 【0†Docs】【2†API Reference】 L5: L6: Search L7: L8: ## Docs L9: L10: ### Getting Started L11: L12: 【0†Overview】【3†Quickstart】【4†Models】【5†OpenAI Compatibility】【6†Responses API】【7†Rate Limits】【8†Templates】【2†API Reference】 L13: L14: ### Core Features L15: L16: 【9†Text Generation】【10†Speech to Text】【11†Text to Speech】【12†Orpheus】【13†OCR and Image Recognition】【14†Reasoning】【15†Content Moderation】【16†Structured Outputs】【17†Prompt Caching】 L17: ### Tools & Integrations L18: L19: 【18†Tool Use】【18†Overview】【19†Groq Built-In Tools】【20†Code Execution】【21†Browser Search (GPT OSS Models)】【22†Remote Tools and MCP】【23†Connectors】【24†Local Tool Calling】【25†Integrations Catalog】【26†Coding with Groq】【27†Factory Droid】【28†OpenCode】【29†Kilo Code】【30†Roo Code】【31†Cline】 L20: L21: ### Guides L22: L23: 【32†Prompting Guide】【32†Basics】【33†Patterns】【34†Model Migration】【35†Assistant Message Prefilling】 L24: ### Service Tiers L25: L26: 【36†Service Tiers】【37†Performance Tier】【38†Flex Processing】【39†Batch Processing】 L27: L28: ### Advanced L29: L30: 【40†LoRA Inference】 L31: L32: ### Production Readiness L33: L34: 【41†Production Checklist】【42†Optimizing Latency】【43†Security Onboarding】【44†Prometheus Metrics】 L35: L36: ### Account and Console L37: L38: 【45†Spend Limits】【46†Projects】【47†Model Permissions】【48†Billing FAQs】【49†Your Data】 L39: ### Developer Resources L40: L41: 【50†SDK Libraries】【51†Groq Badge】【52†Developer Community†community.groq.com】【53†OpenBench†openbench.dev】【54†Error Codes】【55†Changelog】 L42: L43: ### Legal L44: L45: 【56†Policies & Notices】 L46: L47: Search L48: L49: 【0†Docs】【2†API Reference】 L50: L51: ### Getting Started L52: L53: 【0†Overview】 L54: L55: 【3†Quickstart】 L56: L57: 【4†Models】 L58: L59: 【5†OpenAI Compatibility】 L60: L61: 【7†Rate Limits】 L62: L63: 【8†Templates】 L64: L65: 【2†API Reference】 L66: ### Core Features L67: L68: 【9†Text Generation】 L69: L70: 【10†Speech to Text】 L71: L72: 【11†Text to Speech】 L73: L74: 【13†OCR and Image Recognition】 L75: L76: 【14†Reasoning】 L77: L78: 【15†Content Moderation】 L79: L80: 【16†Structured Outputs】 L81: L82: 【17†Prompt Caching】 L83: L84: ### Tools & Integrations L85: L86: 【18†Tool Use】 L87: L88: 【18†Overview】 L89: L90: 【19†Groq Built-In Tools】 L91: L92: 【22†Remote Tools and MCP】 L93: L94: 【24†Local Tool Calling】 L95: L96: 【25†Integrations Catalog】 L97: L98: 【26†Coding with Groq】 L99: L100: ### Guides L101: L102: 【32†Prompting Guide】 L103: L104: 【35†Assistant Message Prefilling】 L105: ### Service Tiers L106: L107: 【36†Service Tiers】 L108: L109: 【37†Performance Tier】 L110: L111: 【38†Flex Processing】 L112: L113: 【39†Batch Processing】 L114: L115: ### Advanced L116: L117: 【40†LoRA Inference】 L118: L119: ### Production Readiness L120: L121: 【41†Production Checklist】 L122: L123: 【42†Optimizing Latency】 L124: L125: 【43†Security Onboarding】 L126: L127: 【44†Prometheus Metrics】 L128: L129: ### Account and Console L130: L131: 【45†Spend Limits】 L132: L133: 【46†Projects】 L134: L135: 【47†Model Permissions】 L136: L137: 【48†Billing FAQs】 L138: L139: 【49†Your Data】 L140: ### Developer Resources L141: L142: 【50†SDK Libraries】 L143: L144: 【51†Groq Badge】 L145: L146: 【52†Developer Community†community.groq.com】 L147: L148: 【53†OpenBench†openbench.dev】 L149: L150: 【54†Error Codes】 L151: L152: 【55†Changelog】 L153: L154: ### Legal L155: L156: 【56†Policies & Notices】 L157: L158: # Your Data in GroqCloud L159: L160: [Button: Copy page] L161: L162: Understand how Groq uses customer data and the controls you have. L163: ## 【57†What Data Groq Retains】 L164: L165: Groq handles two distinct types of information: L166: L167: 1. Usage Metadata (always retained) L168: L169: * We collect usage metadata for all users to measure service activity and system performance. L170: * This metadata does not contain customer inputs or outputs. L171: 2. Customer Data (retained only in limited circumstances) L172: * By default, Groq does not retain customer data for inference requests. L173: * Customer data (inputs, outputs, and related state) is only retained in two cases: L174: 1. If you use features that require data retention to function (e.g., batch jobs, fine-tuning and LoRAs). L175: 2. If needed to protect platform reliability (e.g., to troubleshoot system failures or investigate abuse). L176: L177: You can control these settings yourself in the 【58†Data Controls settings】. L178: ## 【59†When Customer Data May Be Retained】 L179: L180: Review the 【60†Data Location】 section below to learn where data is retained. L181: ### 【61†1. Application State】 L182: L183: Certain API features require data retention to function: L184: L185: * Batch Processing (`/openai/v1/batches`): Input and output files retained for 30 days unless deleted earlier by the customer. L186: * Fine-tuning (`/openai/v1/fine_tunings`): Model weights and training datasets retained until deleted by the customer. L187: L188: To prevent data retention for application state, you can disable these features for all users in your organization in 【58†Data Controls settings】. L189: ### 【62†2. System Reliability and Abuse Monitoring】 L190: L191: As noted above, inference requests are not retained by default. We may temporarily log inputs and outputs only when: L192: L193: * Troubleshooting errors that degrade platform reliability, or L194: * Investigating suspected abuse (e.g. rate-limit circumvention). L195: These logs are retained for up to 30 days, unless legally required to retain longer. You may opt out of this storage in 【58†Data Controls settings】, but you remain responsible for ensuring safe, compliant usage of the services in accordance with 【63†the terms†groq.com】 and 【64†Acceptable Use & Responsible AI Policy】. L196: ## 【65†Summary Table】 L197: Product | Endpoints | Data Retention Type | Retention Period | ZDR Eligible L198: --- | --- | --- | --- | --- L199: Inference | `/openai/v1/chat/completions` L200: `/openai/v1/responses` L201: `/openai/v1/audio/transcriptions` L202: `/openai/v1/audio/translations` L203: `/openai/v1/audio/speech` | System reliability and abuse monitoring | Up to 30 days | Yes L204: Batch | `/openai/v1/batches` L205: `/openai/v1/files` (purpose: `batch`) | Application state | Up to 30 days | Yes (feature disabled) L206: Fine-tuning | `/openai/v1/fine_tunings` L207: `/openai/v1/files` (purpose: `fine_tuning`) | Application state | Until deleted | Yes (feature disabled) L208: ## 【66†Zero Data Retention】 L209: L210: All customers may enable Zero Data Retention (ZDR) in 【58†Data Controls settings】. When ZDR is enabled, Groq will not retain customer data for system reliability and abuse monitoring. As noted above, this also means that features that rely on data retention to function will be disabled. Organization admins can decide to enable ZDR globally or on a per-feature basis at any time on the Data Controls page in 【58†Data Controls settings】. L211: ## 【60†Data Location】 L212: L213: All customer data is retained in Google Cloud Platform (GCP) buckets located in the United States. Groq maintains strict access controls and security standards as detailed in the 【67†Groq Trust Center†trust.groq.com】. Where applicable, Customers can rely on standard contractual clauses (SCCs) for transfers between third countries and the U.S. L214: ## 【68†Key Takeaways】 L215: L216: * Usage metadata: always collected, never includes customer data. L217: * Customer data: not retained by default. Only retained if you opt into persistence features, or in cases for system reliability and abuse monitoring. L218: * Controls: You can manage data retention in 【58†Data Controls settings】, including opting into Zero Data Retention. L219: L220: ### Was this page helpful? L221: L222: Yes No Suggest Edits