Skip to content

Commit 6c5c1c1

Browse files
djw-mdjw-m
authored
Merge pull request #6790 from EnterpriseDB/release-2025-05-12a
Release 2025-05-12a
2 parents 3fd68cb + 0ab2645 commit 6c5c1c1

File tree

5 files changed

+145
-33
lines changed

5 files changed

+145
-33
lines changed

‎advocacy_docs/edb-postgres-ai/ai-accelerator/reference/index.mdx

+2
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,8 @@ navigation:
2121
* [aidb.create_model](models#aidbcreate_model)
2222
* [aidb.get_model](models#aidbget_model)
2323
* [aidb.delete_model](models#aidbdelete_model)
24+
* [aidb.get_hcp_models](models#aidbget_hcp_models)
25+
* [aidb.create_hcp_model](models#aidbcreate_hcp_model)
2426
* [aidb.encode_text](models#aidbencode_text)
2527
* [aidb.encode_text_batch](models#aidbencode_text_batch)
2628
* [aidb.decode_text](models#aidbdecode_text)

‎advocacy_docs/edb-postgres-ai/ai-accelerator/reference/models.mdx

+67-33
Original file line numberDiff line numberDiff line change
@@ -13,8 +13,8 @@ This reference documentation for EDB Postgres AI - AI Accelerator Pipelines mode
1313

1414
The `aidb.model_providers` table stores information about the model providers that are available.
1515

16-
| Column | Type | Description |
17-
|----------------------|----------|----------------------------------|
16+
| Column | Type | Description |
17+
| -------------------- | -------- | ------------------------------- |
1818
| `server_name` | name | Name for the model server |
1919
| `server_description` | text | Description of the model server |
2020
| `server_options` | text\[\] | Options for the model server |
@@ -25,8 +25,8 @@ The `aidb.model_providers` table stores information about the model providers th
2525
Returns a list of all models in the registry and their configured options, including predefined models and user-created models.
2626

2727

28-
| Column | Type | Description |
29-
|------------|-------|------------------------------------------------|
28+
| Column | Type | Description |
29+
| ---------- | ----- | --------------------------------------------- |
3030
| `name` | text | User-defined name for the model |
3131
| `provider` | text | Name of the model provider |
3232
| `options` | jsonb | Optional configuration for the model provider |
@@ -54,13 +54,13 @@ Creates a new model in the system by saving its name, provider, and optional con
5454

5555
#### Parameters
5656

57-
| Parameter | Type | Default | Description |
58-
|-----------------------|---------|-------------|-------------------------------------------------------------------------------------------------------------|
59-
| `name` | text | | User-defined name for the model. |
60-
| `provider` | text | | Name of the model provider (as found in [aidb.model_providers](#aidbmodel_providers)). |
61-
| `config` | jsonb | '{}'::jsonb | Optional configuration for the model provider. May include model-specific parameters such as `model`, `url`, and TLS options (e.g., `tls_config`).|
62-
| `credentials` | jsonb | '{}'::jsonb | Optional credentials for the model provider. |
63-
| `replace_credentials` | boolean | false | If true, replaces the credentials for the model provider. If false, the credentials aren't overwritten. |
57+
| Parameter | Type | Default | Description |
58+
| --------------------- | ------- | ----------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
59+
| `name` | text | | User-defined name for the model. |
60+
| `provider` | text | | Name of the model provider (as found in [aidb.model_providers](#aidbmodel_providers)). |
61+
| `config` | jsonb | '{}'::jsonb | Optional configuration for the model provider. May include model-specific parameters such as `model`, `url`, and TLS options (e.g., `tls_config`). |
62+
| `credentials` | jsonb | '{}'::jsonb | Optional credentials for the model provider. |
63+
| `replace_credentials` | boolean | false | If true, replaces the credentials for the model provider. If false, the credentials aren't overwritten. |
6464

6565

6666
#### Example
@@ -110,14 +110,14 @@ Returns the configuration for a model in the registry.
110110

111111
#### Parameters
112112

113-
| Parameter | Type | Default | Description |
114-
|--------------|------|---------|--------------------|
113+
| Parameter | Type | Default | Description |
114+
| ------------ | ---- | ------- | ----------------- |
115115
| `model_name` | text | | Name of the model |
116116

117117
#### Returns
118118

119-
| Column | Type | Description |
120-
|------------|-------|------------------------------------------------|
119+
| Column | Type | Description |
120+
| ---------- | ----- | --------------------------------------------- |
121121
| `name` | text | User-defined name for the model |
122122
| `provider` | text | Name of the model provider |
123123
| `options` | jsonb | Optional configuration for the model provider |
@@ -139,8 +139,8 @@ Deletes a model from the registry.
139139

140140
#### Parameters
141141

142-
| Parameter | Type | Default | Description |
143-
|--------------|------|---------|--------------------|
142+
| Parameter | Type | Default | Description |
143+
| ------------ | ---- | ------- | ----------------- |
144144
| `model_name` | text | | Name of the model |
145145

146146
#### Example
@@ -156,18 +156,52 @@ __OUTPUT__
156156

157157
#### Returns
158158

159-
| Column | Type | Description |
160-
|---------------------------|-------|----------------------------------------------------------|
159+
| Column | Type | Description |
160+
| -------------- | ----- | ---------------------------------------------------- |
161161
| `delete_model` | jsonb | The name, provider, and options of the deleted model |
162162

163+
164+
### `aidb.get_hcp_models`
165+
166+
Gets models running on the hybrid control plane.
167+
168+
#### Returns
169+
170+
| Column | Type | Description |
171+
| ------- | ---- | ------------------------------------------------- |
172+
| `name` | text | The name of the model instance running on the HCP |
173+
| `url` | text | The API URL of the model running on the HCP |
174+
| `model` | text | The name the model running on the HCP |
175+
176+
#### Example
177+
```sql
178+
SELECT * FROM aidb.get_hcp_models();
179+
name | url | model
180+
---------------------------------------+----------------------------------------------------------------------------------+------------------------------------
181+
llama-3-1-8b-instruct-1xgpu-g6 | http://llama-3-1-8b-instruct-1xgpu-g6-predictor.default.svc.cluster.local | meta/llama-3.1-8b-instruct
182+
llama-3-2-nv-embedqa-1b-v2 | http://llama-3-2-nv-embedqa-1b-v2-predictor.default.svc.cluster.local | nvidia/llama-3.2-nv-embedqa-1b-v2
183+
meta-nim-llama3-70b-instruct-8xgpu-g5 | http://meta-nim-llama3-70b-instruct-8xgpu-g5-predictor.default.svc.cluster.local | meta/llama3-70b-instruct
184+
(3 rows)
185+
```
186+
### `aidb.create_hcp_model`
187+
188+
Creates a new model in the system by referencing a running instance in the HCP
189+
190+
#### Parameters
191+
192+
| Parameter | Type | Default | Description |
193+
| ---------------- | ---- | ------- | ----------------------------------------- |
194+
| `name` | text | | User-defined name of the model |
195+
| `hcp_model_name` | text | | Name of the model instance running on HCP |
196+
163197
### `aidb.encode_text`
164198

165199
Encodes text using a model, generating a vector representation of a given text input.
166200

167201
#### Parameters
168202

169-
| Parameter | Type | Default | Description |
170-
|--------------|------|---------|-----------------------------------|
203+
| Parameter | Type | Default | Description |
204+
| ------------ | ---- | ------- | -------------------------------- |
171205
| `model_name` | text | | Name of the model to encode with |
172206
| `text` | text | | Text to encode |
173207

@@ -177,10 +211,10 @@ Encodes a batch of text using a model, generating a vector representation of a g
177211

178212
#### Parameters
179213

180-
| Parameter | Type | Default | Description |
181-
|--------------|----------|---------|-----------------------------------|
214+
| Parameter | Type | Default | Description |
215+
| ------------ | -------- | ------- | -------------------------------- |
182216
| `model_name` | text | | Name of the model to encode with |
183-
| `text` | text\[\] | | Array of text to encode |
217+
| `text` | text\[\] | | Array of text to encode |
184218

185219

186220
### `aidb.decode_text`
@@ -189,15 +223,15 @@ Decodes text using a model, generating a vector representation of a given text i
189223

190224
#### Parameters
191225

192-
| Parameter | Type | Default | Description |
193-
|--------------|------|---------|-----------------------------------|
226+
| Parameter | Type | Default | Description |
227+
| ------------ | ---- | ------- | -------------------------------- |
194228
| `model_name` | text | | Name of the model to decode with |
195229
| `text` | text | | Text to decode |
196230

197231
#### Returns
198232

199233
| Column | Type | Description |
200-
|---------------|------|------------------|
234+
| ------------- | ---- | ---------------- |
201235
| `decode_text` | text | The decoded text |
202236

203237
### `aidb.decode_text_batch`
@@ -207,14 +241,14 @@ Decodes a batch of text using a model, generating a representation of a given te
207241
#### Parameters
208242

209243
| Parameter | Type | Default | Description |
210-
|--------------|----------|---------|----------------------------------|
244+
| ------------ | -------- | ------- | -------------------------------- |
211245
| `model_name` | text | | Name of the model to decode with |
212246
| `text` | text\[\] | | Array of text to decode |
213247

214248
#### Returns
215249

216250
| Column | Type | Description |
217-
|---------------|------|------------------|
251+
| ------------- | ---- | ---------------- |
218252
| `decode_text` | text | The decoded text |
219253

220254
### `aidb.encode_image`
@@ -224,14 +258,14 @@ Encodes an image using a model, generating a vector representation of a given im
224258
#### Parameters
225259

226260
| Parameter | Type | Default | Description |
227-
|--------------|-------|---------|----------------------------------|
261+
| ------------ | ----- | ------- | -------------------------------- |
228262
| `model_name` | text | | Name of the model to encode with |
229263
| `image` | bytea | | Image to encode |
230264

231265
#### Returns
232266

233267
| Column | Type | Description |
234-
|----------------|-------|-------------------|
268+
| -------------- | ----- | ----------------- |
235269
| `encode_image` | bytea | The encoded image |
236270

237271
### `aidb.rerank_text`
@@ -241,15 +275,15 @@ Reranks text using a model, generating a vector representation of a given text i
241275
#### Parameters
242276

243277
| Parameter | Type | Default | Description |
244-
|--------------|----------|---------|-----------------------------------------------|
278+
| ------------ | -------- | ------- | --------------------------------------------- |
245279
| `model_name` | text | | Name of the model to rerank with |
246280
| `query` | text | | Query based on which the input will be ranked |
247281
| `input` | text\[\] | \[\] | Inputs to be ranked |
248282

249283
#### Returns
250284

251285
| Column | Type | Description |
252-
|---------------|------------------|--------------------------------------------|
286+
| ------------- | ---------------- | ------------------------------------------ |
253287
| `text` | text | The text from "input" |
254288
| `logit_score` | double precision | Score/rank of this text |
255289
| `id` | int | index that the text had in the input array |

‎advocacy_docs/edb-postgres-ai/ai-accelerator/rel_notes/ai-accelerator_4.0.1_rel_notes.mdx

+33
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
---
2+
title: AI Accelerator - Pipelines 4.0.1 release notes
3+
navTitle: Version 4.0.1
4+
originalFilePath: advocacy_docs/edb-postgres-ai/ai-accelerator/rel_notes/src/rel_notes_4.0.1.yml
5+
editTarget: originalFilePath
6+
---
7+
8+
Released: 9 May 2025
9+
10+
This is a minor release that includes a few bug fixes and enhancements to the knowledge base pipeline.
11+
12+
## Highlights
13+
14+
- Bug fixes and performance enhancements.
15+
- Simplified model integration for HCP.
16+
17+
## Enhancements
18+
19+
<table class="table w-100"><thead><tr><th>Description</th><th width="10%">Addresses</th></tr></thead><tbody>
20+
<tr><td><details><summary>Bug fixes for knowledge base and preparer pipeline schema handling.</summary><hr/><p>The knowledge base and preparer pipeline now support arbitrary Postgres schemas for source and destination tables/volumes.
21+
A bug prevented users for configuring explicit schemas via Postgres qualified identifiers (<code>schema.name</code>) when referencing source or destination tables/volumes in the create pipeline calls. This bug is now fixed.</p>
22+
</details></td><td></td></tr>
23+
<tr><td><details><summary>Bug fix for knowledge base result accuracy.</summary><hr/><p>A bug in the batch-processing code for knowledge base pipelines would, under certain circumstances, lead to inaccurate results during retrieval.
24+
This bug was introduced in the 4.0.0 release and is now fixed.</p>
25+
</details></td><td></td></tr>
26+
<tr><td><details><summary>Simplified model integration for HCP.</summary><hr/><p>External models, running on the HCP Model Serving infrastructure, can now be listed and integrated into AIDB with new helper functions <code>aidb.get_hcp_models()</code> and <code>aidb.create_hcp_model()</code>.</p>
27+
</details></td><td></td></tr>
28+
<tr><td><details><summary>Performance enhancement for embeddings processing with external models.</summary><hr/><p>When using external models for embeddings (e.g. with the nim_embeddings model adapter), AIDB performs additional API calls in order to probe the model service for the response type.
29+
These calls are now skipped in most situations to reduce the overhead to a minimum when running embedding processing. AIDB will now use actual results to determine the response type.</p>
30+
</details></td><td></td></tr>
31+
</tbody></table>
32+
33+

‎advocacy_docs/edb-postgres-ai/ai-accelerator/rel_notes/index.mdx

+2
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,7 @@ navTitle: Release notes
44
description: Release notes for EDB Postgres AI - AI Accelerator
55
indexCards: none
66
navigation:
7+
- ai-accelerator_4.0.1_rel_notes
78
- ai-accelerator_4.0.0_rel_notes
89
- ai-accelerator_3.0.1_rel_notes
910
- ai-accelerator_2.2.1_rel_notes
@@ -21,6 +22,7 @@ The EDB Postgres AI - AI Accelerator describes the latest version of AI Accelera
2122

2223
| AI Accelerator version | Release Date |
2324
|---|---|
25+
| [4.0.1](./ai-accelerator_4.0.1_rel_notes) | 09 May 2025 |
2426
| [4.0.0](./ai-accelerator_4.0.0_rel_notes) | 05 May 2025 |
2527
| [3.0.1](./ai-accelerator_3.0.1_rel_notes) | 03 Apr 2025 |
2628
| [2.2.1](./ai-accelerator_2.2.1_rel_notes) | 13 Mar 2025 |

‎advocacy_docs/edb-postgres-ai/ai-accelerator/rel_notes/src/rel_notes_4.0.1.yml

+41
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,41 @@
1+
# yaml-language-server: $schema=https://raw.githubusercontent.com/EnterpriseDB/docs/refs/heads/develop/tools/automation/generators/relgen/relnote-schema.json
2+
product: AI Accelerator - Pipelines
3+
version: 4.0.1
4+
date: 9 May 2025
5+
intro: |
6+
This is a minor release that includes a few bug fixes and enhancements to the knowledge base pipeline.
7+
highlights: |
8+
- Bug fixes and performance enhancements.
9+
- Simplified model integration for HCP.
10+
relnotes:
11+
- relnote: Bug fixes for knowledge base and preparer pipeline schema handling.
12+
details: |
13+
The knowledge base and preparer pipeline now support arbitrary Postgres schemas for source and destination tables/volumes.
14+
A bug prevented users for configuring explicit schemas via Postgres qualified identifiers (`schema.name`) when referencing source or destination tables/volumes in the create pipeline calls. This bug is now fixed.
15+
jira: ""
16+
addresses: ""
17+
type: Enhancement
18+
impact: Medium
19+
- relnote: Bug fix for knowledge base result accuracy.
20+
details: |
21+
A bug in the batch-processing code for knowledge base pipelines would, under certain circumstances, lead to inaccurate results during retrieval.
22+
This bug was introduced in the 4.0.0 release and is now fixed.
23+
jira: "AID-425"
24+
addresses: ""
25+
type: Enhancement
26+
impact: Medium
27+
- relnote: Simplified model integration for HCP.
28+
details: |
29+
External models, running on the HCP Model Serving infrastructure, can now be listed and integrated into AIDB with new helper functions `aidb.get_hcp_models()` and `aidb.create_hcp_model()`.
30+
jira: "AID-416"
31+
addresses: ""
32+
type: Enhancement
33+
impact: Medium
34+
- relnote: Performance enhancement for embeddings processing with external models.
35+
details: |
36+
When using external models for embeddings (e.g. with the nim_embeddings model adapter), AIDB performs additional API calls in order to probe the model service for the response type.
37+
These calls are now skipped in most situations to reduce the overhead to a minimum when running embedding processing. AIDB will now use actual results to determine the response type.
38+
jira: "AID-424"
39+
addresses: ""
40+
type: Enhancement
41+
impact: Medium

0 commit comments

Comments
 (0)