IBM watsonx.ai Runtime
A recurring no-cost API allowance covering 4 cataloged models.
https://{region}.ml.cloud.ibm.com/ml/v1Models mentioned
granite-4-h-smallgranite-4-h-tinygranite-4-h-microgranite-3-1-8b-baseEquivalent paid value
How this was valued: The Lite plan's three independently documented monthly allowances are priced at IBM's current US rates. The 300K shared foundation-token pool is valued once against Granite 4H Small by assigning the envelope to costlier output tokens; other models and regional price differences are excluded.
This is a conservative subtotal because IBM's broader catalog uses model-specific token multipliers and regional prices.
Limits and terms
- Foundation Tokens Per Month
- 300,000
- Requests Per Second
- 2
- Capacity Unit Hours
- 20
- Extraction Pages
- 100
- Plan
- Lite
- Inactivity
- A Lite service can be deleted after 30 days of inactivity.
What happens to your prompts?
IBM says unsaved API prompts and outputs are not accessed, logged, stored, trained on, or used for improvement without permission.
- Plan Scope
- IBM watsonx.ai SaaS foundation-model inference; saved assets, prompt sessions, tuning data, and third-party models have additional state and terms.
- Prompt Retention
- IBM states it does not access, log, or store unsaved API prompts or outputs. Content deliberately saved as project assets is stored until the customer deletes it; temporary prompt-session assets can persist for 30 days.
- Response Retention
- Unsaved API outputs receive the same no-log and no-store treatment; saved results follow project and feature lifecycle controls.
- Ordinary Logging
- IBM retains account and service-usage metadata for operations, security, metering, support, and contractual obligations without treating unsaved prompt content as ordinary logs.
- Model Training
- IBM states customer prompts and outputs are not used to train foundation models or improve the service without permission.
- Product Improvement
- Unsaved prompts and outputs are excluded from service or model improvement; separately supplied feedback, tuning data, or expressly authorized data can be governed by their feature terms.
- Human Or Operator Access
- IBM says unsaved API prompts and outputs are not accessed. Saved assets and support cases can be accessed as needed under IBM Cloud security and contractual controls.
- Subprocessors And Routing
- IBM Cloud subprocessors and, where selected, third-party model providers may process data under the service terms and DPA; model-specific terms can add restrictions.
- Deletion Controls
- Customers can delete saved project assets; prompt-session assets expire after 30 days. Contract termination and personal-data rights follow IBM Cloud terms and privacy procedures.
- Caveat
- The strongest no-store statement applies to unsaved API requests. Prompt Lab history, projects, files, deployments, tuning data, support material, and third-party models can create separately retained state.
Governing documents
- Cloud Service Terms
- https://www.ibm.com/support/customer/csol/terms/
- Privacy Statement
- https://www.ibm.com/privacy
- Data Processing Addendum
- https://www.ibm.com/support/customer/csol/terms/?id=i126-7875
Eligibility
- Account Required
- Yes
- Payment Method Required For New Accounts
- Yes
- Caveat
- New IBM Cloud accounts require a card for identity verification even when using Lite services.
Primary sources
Before you build with IBM watsonx.ai Runtime
Read the classification narrowly
This record has current first-party evidence for some zero-cost hosted inference. The label 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 region.
Resolve the live model route
This snapshot records 4 model IDs. Match the exact ID against the provider's live catalog before using it in code; zero-price routes and aliases can rotate while an older documentation page remains online. The recorded base endpoint is https://{region}.ml.cloud.ibm.com/ml/v1, but the provider's current API reference remains authoritative for paths, authentication, and request shape.
Confirm account and billing boundaries
An account is required, so public catalog evidence cannot by itself prove the limits, balance, or billing behavior attached to your workspace. A payment method is required. A zero-cost allowance can still be useful, but protect the account with provider-side budgets or alerts before sending production traffic. The equivalent paid value shown here prices a documented allowance where possible; it is not cash, guaranteed savings, or protection from overage.
Re-read the prompt policy for your plan
This record classifies the reviewed handling as “Private”: IBM says unsaved API prompts and outputs are not accessed, logged, stored, trained on, or used for improvement without permission. Provider policies can distinguish free, paid, enterprise, opted-in, and feature-specific traffic, so verify the governing terms for the exact account and route that will receive your data.
Follow the evidence, then re-check it
The record links 12 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest governing document or live catalog when sources disagree, and submit a correction when a provider changes a material term.
Match the quota to the workload shape
Translate the published allowance into the traffic pattern you actually expect instead of comparing headline totals alone. A daily token pool can look generous while a low requests-per-minute or concurrency ceiling blocks interactive bursts; a high request limit can still fail a long-context job when input, output, or per-request tokens are capped. Separate prompt tokens from generated tokens, include retries and tool calls, and test the largest realistic payload. If the provider documents more than one limit window, the tightest window at your peak load is the practical ceiling. Research, experimental, and community access can also carry eligibility or fair-use constraints that cannot be modeled as a simple number.
Plan fallback without changing the rules
A fallback should preserve more than API syntax. Confirm that the substitute route supports the required modality, context length, streaming behavior, structured output, tools, and safety controls, then compare its prompt-retention and training terms. An OpenAI-compatible request shape does not make providers operationally or contractually equivalent. Decide which errors may trigger a retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. If deterministic output matters, record the model revision and sampling settings; rotating aliases and free-model pools can change behavior even when the endpoint remains available.
Monitor the offer as a dependency
Capture the model ID, response model field, rate-limit headers, usage fields, latency, HTTP status, and any provider request identifier for each test. Watch for authorization failures, quota exhaustion, catalog removal, policy revisions, and dashboard balance changes as separate failure modes. Re-check the provider’s live catalog and governing pages on a schedule proportionate to the workload’s importance, and keep the dated sources that supported your decision. Free capacity is especially suitable for prototypes, evaluations, fallbacks, and bounded workloads when the application can tolerate change; a production dependency still needs observability, an exit path, and an owner responsible for re-verification.
Test one complete request before scaling
Start with the smallest permitted request using the exact credential, model ID, endpoint, and account type you intend to deploy. Record the HTTP status, response headers, usage fields, latency, and any dashboard balance change. Then exercise the failure path: an invalid model, an exhausted quota, or a rate-limit response should fail clearly without silently switching to a paid route. If the provider supports streaming or tool calls, validate those features separately because a free model can expose a narrower capability set than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and avoid sending sensitive data until the observed route matches the reviewed data agreement.