--- title: "Deepgram free inference" description: "$200 for new accounts; it does not recur. See verified limits, model IDs, privacy terms, and primary sources for Deepgram." canonical_url: "https://freeinferencing.com/provider/deepgram/" md_url: "https://freeinferencing.com/provider/deepgram.md" last_updated: "2026-09-19" --- # Deepgram > $200 for new accounts; it does not recur. ## Classification - **Directory:** Current - **Free access:** One-time trial - **Status:** Current trial - **Confidence:** High - **Payment card:** No - **Account:** Required - **Equivalent paid value:** $200.00 once - **API endpoint:** `https://api.deepgram.com/v1` ## Models mentioned No stable model IDs were recorded in this snapshot. ## Limits and terms ```yaml signup_credit_usd: 200 expiry: none recurrence: false coverage: All public model endpoints. ``` ## What happens to your prompts? **Partially private.** Per-request opt-out can exclude training, but participating accounts permit sampled retention and model improvement. Privacy is audited separately from price. Review the current governing terms before sending sensitive or regulated data. ## Data governance and agreements ```yaml data_governance: review_status: reviewed plan_scope: Deepgram hosted speech and voice APIs; Model Improvement Partnership, request opt-out, Voice Agent history, regional, dedicated, and self-hosted modes differ. prompt_retention: Deepgram may retain fractional samples for model improvement and support when an account participates in the Model Improvement Partnership. Requests with mip_opt_out=true retain content only for transaction processing. response_retention: Transcripts and generated audio follow the same program and feature controls; Voice Agent history can be disabled, while reusable configurations and other customer resources persist. ordinary_logging: Usage logs and request metadata are retained for up to 90 days in the console; summarized usage can persist longer. model_training: Only data contractually included through the voluntary Model Improvement Partnership is used in future model training. The per-request mip_opt_out parameter excludes data from the program. product_improvement: Participating content can improve general speech models and support; opted-out content is processed only for the request. Aggregate service metrics remain. human_or_operator_access: Program data can be accessed for labeling, model development, and support under RBAC, MFA, VPN, and encryption controls; opted-out content is transient. subprocessors_and_routing: Deepgram hosts its own models, including its hosted Whisper implementation, and offers regional, dedicated, and self-hosted endpoints; Whisper requests are not sent to OpenAI. deletion_controls: Set mip_opt_out=true on every applicable API request for transient-only content handling and disable Voice Agent history where needed. Dedicated and self-hosted arrangements provide stronger customer control. caveat: Participation and contract defaults must be checked at the account level; the safe API posture requires sending the opt-out parameter consistently on every request. agreements: terms_of_service: https://deepgram.com/terms privacy_policy: https://deepgram.com/privacy master_services_agreement: https://static.deepgram.com/business/MSA_20240315.pdf ``` ## Eligibility ```yaml account_required: true payment_method_required: false ``` ## API compatibility and modalities - **Compatibility:** Deepgram Rest And Websocket - **Modalities:** Speech To Text, Text To Speech, Voice Agents, Speech / audio ## Before you build with Deepgram ### Read the classification narrowly This record describes the bounded offer supported by evidence on 2026-09-19; it does not guarantee permanence, production suitability, uptime, latency, model quality, or access from every account and region. “One-time trial” should be interpreted together with the Current directory placement, High confidence, No payment-card status, and the plan-specific sources below. ### Resolve the live model route No stable model ID is recorded. Resolve the current machine-readable ID, endpoint, authentication method, and request shape from the provider's live documentation before writing fixed production configuration. ### Match the quota to the workload shape Translate the allowance into peak requests per minute, input and output tokens, concurrency, retries, and every daily or monthly ceiling that applies to the intended account. The first limit reached by the workload is the practical ceiling. A large token pool can still fail interactive bursts, and a high request limit can still fail long-context work. Include tool calls and retry traffic, test the largest realistic payload, and treat research, experimental, and community access as conditional on their eligibility and fair-use terms. ### Preserve billing and privacy boundaries The recorded payment-card requirement is No, and account access is required. Confirm both in the live signup flow, set provider-side budgets when charges are possible, and observe the balance or usage fields after a complete request. The prompt-handling classification is **Partially private** because Per-request opt-out can exclude training, but participating accounts permit sampled retention and model improvement. Re-read the terms for the exact route and plan before sending sensitive, regulated, or proprietary content. ### Follow the evidence, then re-check it This record links 9 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest and most specific governing document or live catalog when sources disagree. A dated finding can become stale even when the page remains online, so retain the source and snapshot that supported the decision and submit a correction when a provider changes a material term. ### Plan fallback without policy drift A fallback should preserve modality, context length, streaming, structured output, tools, safety controls, and data terms, not only API syntax. Decide which errors may retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. Rotating aliases and free pools can change behavior without changing the endpoint, so retain the response model field and test at least one substitute route before the primary offer becomes unavailable. ### Monitor the offer as a dependency Capture the model ID, response model, rate-limit headers, usage fields, latency, HTTP status, and provider request identifier. Watch authorization failure, quota exhaustion, catalog removal, policy revision, and balance movement as separate failure modes. Re-check the live catalog and governing sources on a schedule proportionate to the workload's importance, and keep an owner and exit path for any production dependency on volatile free capacity. ### Test one complete request before scaling Start with the smallest permitted request using the exact credential, model ID, endpoint, region, and account type intended for deployment. Record the status, headers, usage fields, response model, latency, and dashboard balance movement. Then exercise an invalid model, quota exhaustion, or rate limit so failure is explicit and cannot silently switch to a paid route. Validate streaming, structured output, and tool calls separately because a free model can expose fewer features than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and do not send sensitive data until the observed route matches the reviewed agreement. ## Primary sources - [Pricing](https://deepgram.com/pricing): Official Pricing - [Promotional credit expiry](https://developers.deepgram.com/guides/deep-dives/managing-projects): Official Docs - [Deepgram Terms](https://deepgram.com/terms): Official Terms - [Deepgram Privacy Policy](https://deepgram.com/privacy): Official Privacy - [Deepgram Master Services Agreement](https://static.deepgram.com/business/MSA_20240315.pdf): Official Terms - [Deepgram Model Improvement Partnership](https://developers.deepgram.com/docs/the-deepgram-model-improvement-partnership-program): Official Docs - [Deepgram usage-log retention](https://deepgram.com/changelog/deepgram-log-usage-data-limited-to-90-days): Official Docs - [Deepgram accelerates into 2025 empowering 200,000+ developers (2025-01-29)](https://deepgram.com/learn/deepgram-accelerates-into-2025): Official Press Release - [PyPI download stats for deepgram-sdk](https://pypistats.org/packages/deepgram-sdk): Package Registry Stats ## Sitemap See the full [semantic sitemap](/sitemap.md) for every page and markdown mirror.