inference_providers


Description

{“what”:“Registry of AI inference providers with access, compliance, and rate-limit attributes.”,“why”:“Central catalog to manage provider credentials and routing decisions; without it, no single source of truth.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.053978873+00:00”}

Columns

Column Type Size Nulls Auto Default Children Parents Comments
id text 2147483647 null
inference_metrics.model_id Implied Constraint R
inference_offerings.benchmark_run_id Implied Constraint R
inference_offerings.model_id Implied Constraint R
inference_offerings.provider_id inference_offerings_provider_id_fkey R
latency_benchmarks.model_id Implied Constraint R
model_context_windows.model_id Implied Constraint R
quality_scores.model_id Implied Constraint R
rate_limit_tiers.model_id Implied Constraint R

{“what”:“Unique provider identifier, e.g., ‘openai’; used as foreign key in other tables.”,“why”:“Without a stable key, provider references from usage logs or configs become ambiguous.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.066767293+00:00”}

name text 2147483647 null

{“what”:“Human-readable provider name, e.g., ‘OpenAI’.”,“why”:“Display name for UI and reports; id is opaque.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.119707800+00:00”}

website text 2147483647 √ null

{“what”:“Provider’s official website URL.”,“why”:“Link for verification and documentation; missing leaves no reference.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.142057438+00:00”}

hq_country_code text 2147483647 √ null

{“what”:“Two-letter ISO country code of provider headquarters.”,“why”:“Used for legal/compliance assessments; unknown location blocks jurisdiction analysis.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.175498303+00:00”}

jurisdiction_id text 2147483647 √ null

{“what”:“Identifier of the legal jurisdiction governing the provider, e.g., ‘US-CA’.”,“why”:“Determines applicable data protection laws; without it, compliance decisions are guesswork.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.206175231+00:00”}

cloud_act_exposure bool 1 √ false

{“what”:“Flag indicating provider is subject to US CLOUD Act.”,“why”:“Affects data sovereignty decisions; unknown exposure risks non-compliance.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.235698878+00:00”}

signup_model text 2147483647 'unknown'::text

{“what”:“How accounts are created: ‘self-serve’, ‘sales’, ‘unknown’.”,“why”:“Determines procurement workflow; unknown model blocks onboarding steps.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.259583497+00:00”}

free_tier bool 1 √ false

{“what”:“Whether provider offers a free usage tier.”,“why”:“Influences cost estimation and trial eligibility; missing flag misleads budgeting.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.275163184+00:00”}

free_tier_details text 2147483647 √ null

{“what”:“Description of free tier limits, e.g., ‘5k requests/month’.”,“why”:“Clarifies free tier constraints; without it, free tier claims are unverifiable.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.306408620+00:00”}

zero_data_retention bool 1 √ false

{“what”:“Flag indicating provider does not retain prompt/response data.”,“why”:“Critical for privacy compliance; unknown retention status risks data exposure.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.331806414+00:00”}

trains_on_user_data bool 1 √ false

{“what”:“Flag indicating provider may use submitted data for model training.”,“why”:“Determines if data usage is acceptable; unknown training practices violate client agreements.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.344233897+00:00”}

dpa_available bool 1 √ false

{“what”:“Whether provider offers a Data Processing Agreement (DPA).”,“why”:“Required for GDPR/B2B contracts; missing DPA blocks legal sign-off.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.365162135+00:00”}

soc2_certified bool 1 √ false

{“what”:“Whether provider holds SOC 2 Type II certification.”,“why”:“Security assessment relies on this; unknown certification undermines vendor risk review.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.398588107+00:00”}

base_url text 2147483647 ''::text

{“what”:“API endpoint base URL, e.g., ‘https://api.openai.com/v1’.”,“why”:“Needed to construct requests; empty default indicates unconfigured provider.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.419168509+00:00”}

api_key_env text 2147483647 ''::text

{“what”:“Environment variable name holding the API key, e.g., ‘OPENAI_API_KEY’.”,“why”:“Avoids storing secrets; missing name prevents runtime authentication.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.434821664+00:00”}

aggregator bool 1 true

{“what”:“Flag indicating provider is an aggregator (e.g., proxy) rather than direct.”,“why”:“Changes rate limit handling and cost calculation; unknown type misroutes traffic.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.454902146+00:00”}

default_rpm int4 10 √ null

{“what”:“Default requests per minute limit.”,“why”:“Used for throttling; missing value leads to overloading provider.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.474475816+00:00”}

default_tpm int4 10 √ null

{“what”:“Default tokens per minute limit.”,“why”:“Used for token-based throttling; missing value risks 429 errors.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.514713325+00:00”}

default_rpd int4 10 √ null

{“what”:“Default requests per day limit.”,“why”:“Daily quota enforcement; missing value allows exceeding daily caps.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.604061869+00:00”}

default_tpd int4 10 √ null

{“what”:“Default tokens per day limit.”,“why”:“Daily token budget; missing value causes unexpected cost overruns.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.653084863+00:00”}

concurrent_requests int4 10 √ null

{“what”:“Maximum simultaneous requests allowed.”,“why”:“Controls connection pool size; missing value may cause connection saturation.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.712858689+00:00”}

rate_limit_source text 2147483647 √ null

{“what”:“Origin of rate limit data, e.g., ‘provider_docs’ or ‘measured’.”,“why”:“Indicates reliability of limits; unknown source undermines capacity planning.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.783874976+00:00”}

rate_limit_verified_at timestamptz 35,6 √ null

{“what”:“Timestamp when rate limits were last verified.”,“why”:“Stale limits mislead capacity; missing timestamp hides outdated data.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.846192343+00:00”}

enabled bool 1 true

{“what”:“Flag indicating provider is active for use.”,“why”:“Disabled providers must be excluded from routing; missing flag could send traffic to dead endpoints.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.883403924+00:00”}

notes text 2147483647 √ null

{“what”:“Free-form internal notes about the provider.”,“why”:“Captures context not in schema; missing notes lose operational knowledge.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.932557974+00:00”}

created_at timestamptz 35,6 now()

{“what”:“Timestamp when provider record was created.”,“why”:“Audit trail and ordering; missing creation time complicates lifecycle tracking.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:24.949771942+00:00”}

updated_at timestamptz 35,6 now()

{“what”:“Timestamp of last update to any field.”,“why”:“Detects stale configuration; missing update time hides recent changes.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:14:25.032750772+00:00”}

Indexes

Constraint Name Type Sort Column(s)
inference_providers_pkey Primary key Asc id

Relationships