Models reference

Reference for all model-related functions and views in AIDB. For guide-style documentation, see Integrating models .

Configuration model

A model’s behavior is controlled two ways, and it matters which one you use for a given setting:

• Creation-time ``config`` — set when you call aidb.create_model() , using a provider-specific Model config helpers (aidb.bert_config() , aidb.completions_config() , and so on). This is the model’s persistent, stored configuration.

• Call-time ``inference_config`` — an optional per-call override, built with aidb.inference_config() , passed directly to an inference function. It overrides the model’s stored config for that one call only; anything you don’t override falls back to the value set at creation time.

Only some inference functions accept a call-time override — the rest are fixed entirely by the model’s creation-time config :

Function	Call-time override?
aidb.generate_text(), aidb.generate_text_batch()	Yes — pass aidb.inference_config() (or raw JSON) as the inference_config argument.
aidb.summarize_text(), aidb.summarize_text_aggregate()	Yes — via the inference_config field of AI pipelines .
aidb.encode_text(), aidb.encode_text_batch(), aidb.encode_image()	No. Behavior is fixed by config at creation time.
aidb.rerank_text()	No. Behavior is fixed by config at creation time.
aidb.perform_ocr()	No — its options parameter only selects which registered model to run, not inference settings.

This asymmetry follows from what each capability needs: text generation has runtime knobs worth tuning per call (temperature, system prompt, and so on); embeddings, reranking, and OCR don’t, so there’s nothing to override.

Catalog views

aidb.model_providers

Lists all available model providers registered in the system.

Column	Type	Description
server_name	name	Name of the model provider
server_description	text	Description of the provider
server_options	text[]	Available configuration options

aidb.models

Lists all models in the registry, including default and user-created models.

Column	Type	Description
name	text	User-defined name for the model
provider	text	Model provider name
options	text[]	Configured options for the model
functions	text[]	Internal capability identifiers the model's provider supports (for example, aidb-openai-text-embeddings, aidb-openai-text-completion)

Example

SELECT * FROM aidb.models;

__OUTPUT__
  name  |  provider  |    options    |                      functions
- -------+------------+---------------+-------------------------------------------------------
 bert   | bert_local | {"config={}"} | {aidb-openai-text-embeddings}
 clip   | clip_local | {"config={}"} | {aidb-openai-text-image-embeddings}
 t5     | t5_local   | {"config={}"} | {aidb-openai-text-embeddings,aidb-openai-text-completion}
(3 rows)

---

Model management functions

aidb.create_model

Registers a new model in the AIDB model registry.

Parameters

Parameter	Type	Default	Description
name	TEXT	Required	Unique name for the model.
provider	TEXT	Required	Provider name (see aidb.model_providers).
config	JSONB	'{}'	Provider-specific configuration. Build with a Model config helpers .
credentials	JSONB	'{}'	Provider credentials (for example, {"api_key": "..."}).
replace_credentials	BOOLEAN	false	If true, updates stored credentials without re-creating the model.
validate	BOOLEAN	true	If true, probe the model at creation time to confirm it's functional. See Validation .
credentials_env	TEXT	NULL	Name of an environment variable to read credentials from instead of storing them. See Credentials from environment variables . Mutually exclusive with credentials.

Credentials from environment variables

!!!note Don’t want a provider’s API key or password sitting in a database table at all, even one with restricted grants? Pass credentials_env instead of credentials , naming a Postgres environment variable to read the credential from. Only the variable’s name is stored — the value itself is read fresh from the Postgres backend process’s environment every time the model is used, and is never persisted in the database.

Pass a variable name instead of a credentials object:

SELECT aidb.create_model(
    name            => my_openai,
    provider        => openai_embeddings,
    config          => aidb.embeddings_config(model => text-embedding-3-small),
    credentials_env => AIDB_OPENAI_API_KEY
);

The referenced variable can hold either a bare secret (used as api_key ) or a JSON object (for example, {"basic_auth": "..."} ) — the same shapes credentials itself accepts. credentials and credentials_env are mutually exclusive; passing both raises an error.

For security, the environment variable’s name must start with the prefix configured by aidb.env_var_allowed_prefix (AIDB_ by default) — this stops a model or MCP server configuration from naming an arbitrary, unrelated environment variable on the Postgres host and exfiltrating its value. The same mechanism and the same GUC apply to MCP server authentication; see headers_env .

TLS configuration

To connect to HTTPS model endpoints, include a tls_config field inside config :

"tls_config": {
  "insecure_skip_verify": true,
  "ca_path": "/etc/aidb/myCA.pem"
}

Validation

By default (validate => true ), aidb.create_model() runs a minimal probe inference before the registration commits, to confirm the model is functional:

• Local models are downloaded and loaded, and a test inference is run — so a bad model identifier or unloadable weights fail at creation time rather than on first use.

• External models receive a small test request — so an invalid API key, wrong URL, or unreachable endpoint is caught right away.

If the probe fails, aidb.create_model() raises an error and registers nothing. Pass validate => false to skip the probe — for example, to defer a large download to first use, or to register a model before its endpoint or credentials are available. You can validate a model later with aidb.validate_model() .

Examples

- - Minimal registration
SELECT aidb.create_model(my_t5, t5_local);

- - With config and credentials
SELECT aidb.create_model(
    name        => my_t5,
    provider    => t5_local,
    config      => {"param1": "value1"}::JSONB,
    credentials => {"token": "abcd"}::JSONB
);

- - Rotate credentials without re-creating
SELECT aidb.create_model(
    name                => my_openai,
    provider            => openai_embeddings,
    config              => aidb.embeddings_config(model => text-embedding-3-small),
    credentials         => {"api_key": "<new-key>"}::JSONB,
    replace_credentials => true
);

---

aidb.get_model

Returns the configuration for a registered model.

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the model to retrieve.

Returns

Column	Type	Description
name	text	Model name
provider	text	Provider name
options	text[]	Configured options

Example

SELECT * FROM aidb.get_model(t5);

__OUTPUT__
 name | provider |    options
- -----+----------+---------------
 t5   | t5_local | {"config={}"}
(1 row)

---

aidb.delete_model

Removes a model from the registry. Doesn’t affect pipelines or knowledge bases that reference it until they are next executed.

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the model to delete.

Returns

The name, provider, and options of the deleted model.

Example

SELECT aidb.delete_model(t5);

__OUTPUT__
     delete_model
- --------------------------------
 (t5,t5_local,"{""config={}""}")
(1 row)

---

aidb.validate_model

Probes a registered model to confirm it’s functional, running a minimal test inference. Called automatically by aidb.create_model() unless validate => false ; you can also call it directly to recheck an existing model.

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the model to validate.

Returns

TEXT — the model capability that was probed (for example, text embedding , language , reranking , image embedding , or OCR ). Raises an error if the probe fails.

Example

SELECT aidb.validate_model(my_bert);

__OUTPUT__
 validate_model
- ---------------
 text embedding
(1 row)

---

HCP model functions

These functions manage models running on EDB Hybrid Manager (HM).

aidb.list_hcp_models

Lists models currently running on HM.

Returns

Column	Type	Description
name	text	Model instance name on HCP
url	text	API endpoint URL
model	text	Model identifier

Example

SELECT * FROM aidb.list_hcp_models();

__OUTPUT__
             name              |                         url                          |            model
- ------------------------------+------------------------------------------------------+------------------------------
 llama-3-1-8b-instruct-1xgpu   | http://llama-3-1-8b-predictor.default.svc.local      | meta/llama-3.1-8b-instruct
(1 row)

---

aidb.create_hcp_model

Registers an HCP-hosted model by referencing its running instance name.

Parameters

Parameter	Type	Description
name	TEXT	User-defined name for the model in AIDB.
hcp_model_name	TEXT	Name of the model instance running on HCP.

---

aidb.sync_hcp_models

Synchronizes the AIDB model registry with models currently running on HCP. Creates entries for new HCP models and deletes entries for models no longer running there.

Returns

Column	Type	Description
status	text	created, deleted, unchanged, or skipped
model	text	Name of the synchronized model

---

Inference functions

!!!note aidb.generate_text() and aidb.generate_text_batch() were previously named aidb.decode_text() and aidb.decode_text_batch() . The old names still work but are deprecated and will be removed in a future version — use the new names in new code.

aidb.encode_text

Encodes a single text string into a vector using the specified model.

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the registered model.
input	TEXT	Text to encode.

Returns

real[] — the embedding for the input text. Cast to vector to use with pgvector operators.

Example

SELECT aidb.encode_text(bert_local, The quick brown fox);

---

aidb.encode_text_batch

Encodes an array of text strings into vectors in a single call.

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the registered model.
input	TEXT[]	Array of strings to encode.

Returns

SETOF real[] — one row per input, in input order. The result column is named encode_text_batch .

Example

SELECT * FROM aidb.encode_text_batch(bert_local, ARRAY[The quick brown fox, Lorem ipsum dolor sit amet]);

---

aidb.encode_image

Encodes a binary image into a vector using a multimodal model (for example, CLIP).

Parameters

Parameter	Type	Description
model_name	TEXT	Name of the registered model.
input	BYTEA	Raw image bytes.

Returns

real[] — the embedding for the input image. Cast to vector to use with pgvector operators.

Example

SELECT aidb.encode_image(clip_local, pg_read_binary_file(/tmp/photo.jpg)::BYTEA);

---

aidb.generate_text

Generates a text response from a prompt using the specified model.

Parameters

Parameter	Type	Default	Description
model_name	TEXT	Required	Name of the registered model.
input	TEXT	Required	Text prompt or input.
inference_config	JSON	NULL	Runtime inference settings. Use :ref:``aidb.inference_config()` <#aidbinference_config>` . Cast the result to json.

Returns

TEXT — the generated response.

Examples

- - Basic usage
SELECT aidb.generate_text(t5_local, translate to French: Hello, world.);

- - With inference configuration
SELECT aidb.generate_text(
    my_llama,
    Explain quantum computing in one sentence.,
    aidb.inference_config(
        system_prompt => Be concise and factual.,
        temperature   => 0.7,
        max_tokens    => 100
    )::json
);

---

aidb.generate_text_batch

Generates text responses for an array of prompts in a single call.

Parameters

Parameter	Type	Default	Description
model_name	TEXT	Required	Name of the registered model.
input	TEXT[]	Required	Array of text prompts.
inference_config	JSON	NULL	Runtime inference settings. Use :ref:``aidb.inference_config()` <#aidbinference_config>` . Cast the result to json.

Returns

SETOF TEXT — one row per input prompt, in input order. The result column is named generate_text_batch .

Example

SELECT aidb.generate_text_batch(t5_local, ARRAY[
    translate to German: hello,
    translate to German: goodbye
]);

---

aidb.rerank_text

Scores and ranks a set of text inputs against a query using a reranking model.

Parameters

Parameter	Type	Default	Description
model_name	TEXT	Required	Name of a registered reranking model.
query	TEXT	Required	The query to rank inputs against.
input	TEXT[]	[]	Array of candidate texts to rank.

Returns

Column	Type	Description
text	text	The candidate text
logit_score	double precision	Relevance score (higher = more relevant)
id	int	Original index of the text in the input array

Example

SELECT * FROM aidb.rerank_text(
    my_reranker,
    How do I configure AIDB?,
    ARRAY[
        AIDB requires shared_preload_libraries.,
        Postgres supports JSON natively.,
        Run CREATE EXTENSION aidb CASCADE to install.
    ]
) ORDER BY logit_score DESC;

---

Inference configuration helper

aidb.inference_config

Builds a JSONB configuration object for runtime inference settings. Pass the result (cast to json ) to aidb.generate_text() , aidb.generate_text_batch() , or use it with aidb.summarize_text_config() .

All parameters are optional — omit any you don’t need.

Parameters

Parameter	Type	Default	Description
system_prompt	TEXT	NULL	System prompt prepended to the request. Not supported by T5 models.
temperature	DOUBLE PRECISION	NULL	Sampling temperature. 0.0 is deterministic; higher values increase variety.
max_tokens	INTEGER	NULL	Maximum tokens to generate.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
seed	BIGINT	NULL	Random seed for reproducible outputs.
repeat_penalty	REAL	NULL	Penalty for repeating tokens. 1.0 = no penalty. Not supported by NIM or OpenAI.
repeat_last_n	INTEGER	NULL	Number of recent tokens considered for repeat_penalty. Not supported by NIM or OpenAI.
thinking	BOOLEAN	NULL	Controls <think> tag handling. true/NULL retains tags; false strips them.
extra_args	JSONB	NULL	Additional provider-specific arguments passed directly to the API.

Returns

JSONB — cast to ::json before passing to inference functions.

Examples

- - With temperature and token limit
SELECT aidb.generate_text(
    my_llama,
    What is machine learning?,
    aidb.inference_config(temperature => 0.5, max_tokens => 150)::json
);

- - Full configuration
SELECT aidb.generate_text(
    my_llama,
    Explain relativity.,
    aidb.inference_config(
        system_prompt  => You are a physics professor. Be precise.,
        temperature    => 0.3,
        max_tokens     => 200,
        top_p          => 0.9,
        seed           => 42,
        repeat_penalty => 1.1
    )::json
);

- - Hide reasoning tags from a thinking model
SELECT aidb.generate_text(
    my_reasoning_model,
    Solve: 2x + 5 = 15,
    aidb.inference_config(thinking => false)::json
);

Tool calling and structured output

Three more inference settings exist but aren’t exposed as named parameters of aidb.inference_config() — build them as raw JSON instead, merged with the helper’s output using || if you also need other settings. Only ``llamacpp_generate`` , ``openai_responses`` (+ ``openai_responses_azure`` ), and ``anthropic_messages`` (+ ``anthropic_messages_azure`` /``anthropic_messages_bedrock`` ) honor these; every other provider rejects them.

Field	Type	Description
tools	JSONB array	OpenAI-compatible tools array — function definitions the model can call.
tool_choice	JSONB	OpenAI-compatible tool_choice: the string "auto", "required", or "none", or {"type": "function", "function": {"name": "..."}} to force a specific tool.
response_format	JSONB	Structured output. {"type": "json_schema", "json_schema": {"schema": {...}}} constrains the response to a JSON schema. Mutually exclusive with tools/tool_choice on openai_responses and anthropic_messages (and their variants).

!!!note openai_responses and anthropic_messages (and their Azure/Bedrock variants) map these to each API’s native tools /tool_choice request fields and parse real tool-call responses back — see OpenAI Responses API and Anthropic Messages API . Every other remote provider (openai_completions , gemini , nim_completions , and so on) rejects tools /tool_choice /response_format outright — this rejection only applies to direct calls like aidb.generate_text() . Any of those providers (except t5_local ) can still back an Agents reference ; the agent framework falls back to simulating tool calls by describing them in the prompt and parsing the model’s reply, rather than using these three fields.

- - Tool calling
SELECT aidb.generate_text(
    my_llamacpp_chat,
    What is the weather in Paris?,
    (aidb.inference_config(temperature => 0.2) || jsonb_build_object(
        tools, jsonb_build_array(
            jsonb_build_object(
                type, function,
                function, jsonb_build_object(
                    name, get_weather,
                    description, Get current weather for a city,
                    parameters, jsonb_build_object(
                        type, object,
                        properties, jsonb_build_object(city, jsonb_build_object(type, string)),
                        required, jsonb_build_array(city)
                    )
                )
            )
        ),
        tool_choice, auto
    ))::json
);

- - Structured output
SELECT aidb.generate_text(
    my_llamacpp_chat,
    Extract the city from: I live in Paris.,
    jsonb_build_object(
        response_format, jsonb_build_object(
            type, json_schema,
            json_schema, jsonb_build_object(
                schema, jsonb_build_object(
                    type, object,
                    properties, jsonb_build_object(city, jsonb_build_object(type, string)),
                    required, jsonb_build_array(city)
                )
            )
        )
    )::json
);

---

Model config helpers

These functions return a JSONB configuration object for the config parameter of aidb.create_model() . Each helper corresponds to a specific model provider.

aidb.embeddings_config

For openai_embeddings and any OpenAI-compatible embeddings endpoint.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier.
api_key	TEXT	NULL	API key.
url	TEXT	NULL	Endpoint URL override.
basic_auth	TEXT	NULL	Basic auth (user:password).
max_concurrent_requests	INTEGER	NULL	Max concurrent requests.
max_batch_size	INTEGER	NULL	Max inputs per batch.
input_type	TEXT	NULL	Input type hint (provider-specific).
input_type_query	TEXT	NULL	Query input type hint (provider-specific).
tls_config	JSONB	NULL	TLS configuration.
is_hcp_model	BOOLEAN	NULL	true if model runs on HCP.

Example

SELECT aidb.create_model(
    my_embedder,
    openai_embeddings,
    config => aidb.embeddings_config(
        model   => text-embedding-3-small,
        api_key => sk-...,
        url     => https://api.openai.com/v1
    )
);

---

aidb.completions_config

For openai_completions and any OpenAI-compatible completions endpoint.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier.
api_key	TEXT	NULL	API key.
url	TEXT	NULL	Endpoint URL override.
basic_auth	TEXT	NULL	Basic auth (user:password).
system_prompt	TEXT	NULL	Default system prompt.
temperature	DOUBLE PRECISION	NULL	Sampling temperature.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
seed	BIGINT	NULL	Random seed.
thinking	BOOLEAN	NULL	<think> tag handling.
max_tokens	JSONB	NULL	Max tokens config (use :ref:``aidb.max_tokens_config()` <#aidbmax_tokens_config>` ).
max_concurrent_requests	INTEGER	NULL	Max concurrent requests.
extra_args	JSONB	NULL	Additional provider-specific arguments.
is_hcp_model	BOOLEAN	NULL	true if model runs on HCP.

Example

SELECT aidb.create_model(
    my_llm,
    openai_completions,
    config => aidb.completions_config(
        model       => gpt-4o,
        api_key     => sk-...,
        temperature => 0.2
    )
);

!!!note For agentic use with reliable, native tool calling on OpenAI models, use aidb.openai_responses_config instead — see Tool calling and structured output for how the two providers differ, and Agents reference for building an agent around either one.

---

aidb.max_tokens_config

Builds the max_tokens object for use with aidb.completions_config() .

Parameter	Type	Default	Description
size	INTEGER	Required	Maximum tokens to generate.
format	TEXT	NULL	Format: 'default', 'legacy', or 'both'.

Example

SELECT aidb.completions_config(
    model      => gpt-4o,
    max_tokens => aidb.max_tokens_config(size => 1024, format => default)
);

---

aidb.bert_config

For the bert_local provider.

Parameter	Type	Default	Description
model	TEXT	Required	HuggingFace model identifier.
revision	TEXT	NULL	Model revision or branch.
cache_dir	TEXT	NULL	Local cache directory.

Example

SELECT aidb.create_model(my_bert, bert_local,
    config => aidb.bert_config(sentence-transformers/all-MiniLM-L6-v2));

---

aidb.clip_config

For the clip_local provider.

Parameter	Type	Default	Description
model	TEXT	Required	HuggingFace model identifier.
revision	TEXT	NULL	Model revision or branch.
cache_dir	TEXT	NULL	Local cache directory.
image_size	INTEGER	NULL	Input image size in pixels.

Example

SELECT aidb.create_model(my_clip, clip_local,
    config => aidb.clip_config(openai/clip-vit-base-patch32));

---

aidb.llama_config

For the llama_instruct_local provider.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier or HuggingFace path.
revision	TEXT	NULL	Model revision.
cache_dir	TEXT	NULL	Local cache directory.
model_path	TEXT	NULL	Explicit local path to model weights (overrides model).
system_prompt	TEXT	NULL	Default system prompt.
temperature	DOUBLE PRECISION	NULL	Sampling temperature.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
seed	BIGINT	NULL	Random seed.
sample_len	INTEGER	NULL	Max tokens to generate.
repeat_penalty	REAL	NULL	Repetition penalty.
repeat_last_n	INTEGER	NULL	Tokens considered for repetition penalty.
use_flash_attention	BOOLEAN	NULL	Enable flash attention.
use_kv_cache	BOOLEAN	NULL	Enable KV cache.

Example

SELECT aidb.create_model(my_llama, llama_instruct_local,
    config => aidb.llama_config(
        meta-llama/Llama-3.2-3B-Instruct,
        temperature => 0.5
    )
);

---

aidb.t5_config

For the t5_local provider.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier or HuggingFace path.
revision	TEXT	NULL	Model revision.
model_path	TEXT	NULL	Explicit local path to model weights.
cache_dir	TEXT	NULL	Local cache directory.
temperature	DOUBLE PRECISION	NULL	Sampling temperature.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
seed	BIGINT	NULL	Random seed.
max_tokens	INTEGER	NULL	Max tokens to generate.
repeat_penalty	REAL	NULL	Repetition penalty.
repeat_last_n	INTEGER	NULL	Tokens considered for repetition penalty.

Example

SELECT aidb.create_model(my_t5, t5_local,
    config => aidb.t5_config(google/flan-t5-base));

---

aidb.gemini_config

For the gemini provider.

Parameter	Type	Default	Description
api_key	TEXT	Required	Google API key.
model	TEXT	NULL	Gemini model identifier (for example, gemini-2.0-flash).
url	TEXT	NULL	API endpoint override.
max_concurrent_requests	INTEGER	NULL	Max concurrent requests.
thinking_budget	INTEGER	NULL	Extended thinking token budget (Gemini 2.x only).

Example

SELECT aidb.create_model(my_gemini, gemini,
    config => aidb.gemini_config(AIza..., model => gemini-2.0-flash));

!!!note Unlike other remote providers, Gemini’s api_key is a plain field of config rather than being passed through aidb.create_model() ’s separate credentials parameter. It’s stored as part of the model’s config rather than the more restricted credentials store — keep that in mind if aidb.models.options is visible to more roles than you want to see the key.

---

aidb.nim_clip_config

For the nim_clip provider (NVIDIA NIM multimodal embeddings).

Parameter	Type	Default	Description
api_key	TEXT	NULL	NIM API key.
model	TEXT	NULL	NIM CLIP model identifier.
url	TEXT	NULL	NIM endpoint URL override.
basic_auth	TEXT	NULL	Basic auth credentials.
is_hcp_model	BOOLEAN	NULL	true if model runs on HCP.

Example

SELECT aidb.create_model(my_nim_clip, nim_clip,
    config => aidb.nim_clip_config(api_key => nvapi-..., model => nvidia/nvclip));

---

aidb.nim_ocr_config

For the nim_ocr provider (NVIDIA NIM OCR).

Parameter	Type	Default	Description
api_key	TEXT	NULL	NIM API key.
model	TEXT	NULL	NIM OCR model identifier.
url	TEXT	NULL	NIM endpoint URL override.
basic_auth	TEXT	NULL	Basic auth credentials.
is_hcp_model	BOOLEAN	NULL	true if model runs on HCP.

Example

SELECT aidb.create_model(my_nim_ocr, nim_ocr,
    config => aidb.nim_ocr_config(api_key => nvapi-...));

---

aidb.nim_reranking_config

For the nim_reranking provider.

Parameter	Type	Default	Description
api_key	TEXT	NULL	NIM API key.
model	TEXT	NULL	NIM reranking model identifier.
url	TEXT	NULL	NIM endpoint URL override.
basic_auth	TEXT	NULL	Basic auth credentials.
is_hcp_model	BOOLEAN	NULL	true if model runs on HCP.

Example

SELECT aidb.create_model(my_reranker, nim_reranking,
    config => aidb.nim_reranking_config(
        api_key => nvapi-...,
        model   => nvidia/nv-rerankqa-mistral-4b-v3
    )
);

---

aidb.openrouter_chat_config

For the openrouter_chat provider.

Parameter	Type	Default	Description
model	TEXT	Required	OpenRouter model identifier.
api_key	TEXT	NULL	OpenRouter API key.
url	TEXT	NULL	API endpoint override.
max_concurrent_requests	INTEGER	NULL	Max concurrent requests.
max_tokens	JSONB	NULL	Max tokens config (use :ref:``aidb.max_tokens_config()` <#aidbmax_tokens_config>` ).

Example

SELECT aidb.create_model(my_or_chat, openrouter_chat,
    config => aidb.openrouter_chat_config(
        anthropic/claude-3-5-haiku,
        api_key => sk-or-...
    )
);

---

aidb.openrouter_embeddings_config

For the openrouter_embeddings provider.

Parameter	Type	Default	Description
model	TEXT	Required	OpenRouter embeddings model identifier.
api_key	TEXT	NULL	OpenRouter API key.
url	TEXT	NULL	API endpoint override.
max_concurrent_requests	INTEGER	NULL	Max concurrent requests.
max_batch_size	INTEGER	NULL	Max inputs per batch.

Example

SELECT aidb.create_model(my_or_embedder, openrouter_embeddings,
    config => aidb.openrouter_embeddings_config(
        mistral/mistral-embed,
        api_key => sk-or-...
    )
);

---

aidb.openai_responses_config

For the openai_responses and openai_responses_azure providers.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier (for example, gpt-5.1).
api_key	TEXT	NULL	API key for authentication.
basic_auth	TEXT	NULL	Basic auth credentials (user:password).
url	TEXT	NULL	API endpoint URL. Defaults to OpenAI's Responses endpoint for openai_responses; required for openai_responses_azure (no universal default — use your resource's endpoint).
max_concurrent_requests	INTEGER	25	Maximum concurrent requests to the endpoint.
system_prompt	TEXT	NULL	Default system prompt, sent via the Responses API's native instructions field.
temperature	DOUBLE PRECISION	NULL	Sampling temperature.
max_output_tokens	INTEGER	NULL	Maximum tokens to generate.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
extra_args	JSONB	NULL	Additional provider-specific arguments passed directly to the API.

Examples

- - OpenAI
SELECT aidb.create_model(
    my_gpt,
    openai_responses,
    config      => aidb.openai_responses_config(model => gpt-5.1),
    credentials => {"api_key": "sk-..."}::JSONB
);

- - Azure AI Foundry
SELECT aidb.create_model(
    my_gpt_azure,
    openai_responses_azure,
    config      => aidb.openai_responses_config(
        model => gpt-5.1,
        url   => https://<resource>.openai.azure.com/openai/v1/responses
    ),
    credentials => {"api_key": "..."}::JSONB
);

This provider natively supports the tools , tool_choice , and response_format fields described in Tool calling and structured output .

---

aidb.anthropic_messages_config

For the anthropic_messages , anthropic_messages_azure , and anthropic_messages_bedrock providers.

Parameter	Type	Default	Description
model	TEXT	Required	Model identifier (for example, claude-opus-4-6, or a Bedrock model ID for the Bedrock variant).
api_key	TEXT	NULL	API key for authentication. For anthropic_messages_bedrock, this is a Bedrock API key (bearer token), not an AWS SigV4 credential.
basic_auth	TEXT	NULL	Basic auth credentials (user:password).
url	TEXT	NULL	API endpoint URL. Defaults to api.anthropic.com for anthropic_messages. Required for anthropic_messages_azure (your resource's Messages endpoint) and anthropic_messages_bedrock (the regional bedrock-runtime base endpoint — the model ID is appended automatically).
max_concurrent_requests	INTEGER	25	Maximum concurrent requests to the endpoint.
system_prompt	TEXT	NULL	Default system prompt, sent via the Messages API's native system field.
temperature	DOUBLE PRECISION	NULL	Sampling temperature.
max_tokens	INTEGER	4096	Maximum tokens to generate. Anthropic's Messages API requires this field; AIDB defaults it when omitted.
top_p	DOUBLE PRECISION	NULL	Nucleus sampling threshold.
extra_args	JSONB	NULL	Additional provider-specific arguments passed directly to the API.

Examples

- - Anthropic direct
SELECT aidb.create_model(
    my_claude,
    anthropic_messages,
    config      => aidb.anthropic_messages_config(model => claude-opus-4-6),
    credentials => {"api_key": "sk-ant-..."}::JSONB
);

- - Azure AI Foundry
SELECT aidb.create_model(
    my_claude_azure,
    anthropic_messages_azure,
    config      => aidb.anthropic_messages_config(
        model => claude-opus-4-6,
        url   => https://<resource>.services.ai.azure.com/anthropic/v1/messages
    ),
    credentials => {"api_key": "..."}::JSONB
);

- - AWS Bedrock
SELECT aidb.create_model(
    my_claude_bedrock,
    anthropic_messages_bedrock,
    config      => aidb.anthropic_messages_config(
        model => anthropic.claude-opus-4-6-v1:0,
        url   => https://bedrock-runtime.us-east-1.amazonaws.com
    ),
    credentials => {"api_key": "..."}::JSONB
);

This provider natively supports the tools , tool_choice , and response_format fields described in Tool calling and structured output . Anthropic’s Messages API has no native structured-output field, so response_format is implemented as a forced call to a synthetic tool — functionally equivalent, but it means response_format and tools /tool_choice can’t be used in the same call.