Build(deps-dev): Bump rack from 3.1.16 to 3.1.18

Bumps [rack](https://github.com/rack/rack) from 3.1.16 to 3.1.18. - [Release notes](https://github.com/rack/rack/releases) - [Changelog](https://github.com/rack/rack/blob/main/CHANGELOG.md) - [Commits](https://github.com/rack/rack/compare/v3.1.16...v3.1.18) --- updated-dependencies: - dependency-name: rack dependency-version: 3.1.18 dependency-type: indirect ... Signed-off-by: dependabot[bot] <[email protected]>
DEV: Mention the core move in readme (#1514 )
2025-12-27 22:27:24 +00:00 · 2025-07-22 11:08:32 +01:00 · 2025-07-22 11:05:52 +01:00 · 2025-07-22 16:46:56 +08:00 · 2025-07-22 10:09:03 +08:00 · 2025-07-21 15:36:39 -03:00
281 changed files with 1107 additions and 946 deletions
@@ -1,3 +1,4 @@
+< 3.5.0.beta8-dev: 5d80a34589d63e381d80273828072badc18955b1
 < 3.5.0.beta7-dev: cd14b0c0bee0cf63c59b64b6f7213e31a37f11a7
 < 3.5.0.beta6-dev: 3e74eea1e5e3143888d67a8d8a11206df214dc24
 < 3.5.0.beta3-dev: 09a68414804a1447f52e5d60691ba59742cda9ec
@@ -34,7 +34,7 @@ GEM
      racc
    prettier_print (1.2.1)
    racc (1.8.1)
-    rack (3.1.16)
+    rack (3.1.18)
    rainbow (3.1.1)
    regexp_parser (2.10.0)
    rubocop (1.74.0)
@@ -1,15 +1,18 @@
 # **Discourse AI** Plugin

+> [!IMPORTANT]
+> This plugin has now been bundled into Discourse core. See: https://meta.discourse.org/t/bundling-more-popular-plugins-with-discourse-core/373574
+
 **Plugin Summary**

 For more information, please see: https://meta.discourse.org/t/discourse-ai/259214?u=falco

 ### Evals

-The directory `evals` contains AI evals for the Discourse AI plugin. 
+The directory `evals` contains AI evals for the Discourse AI plugin.
 You may create a local config by copying `config/eval-llms.yml` to `config/eval-llms.local.yml` and modifying the values.

-To run them use: 
+To run them use:

 cd evals
 ./run --help
@@ -40,7 +40,7 @@ module Jobs
      return if locales.blank?

      locales.each do |locale|
-        next if DiscourseAi::Translation::LocaleNormalizer.is_same?(locale, detected_locale)
+        next if LocaleNormalizer.is_same?(locale, detected_locale)
        regionless_locale = locale.split("_").first
        next if post.post_localizations.where("locale LIKE ?", "#{regionless_locale}%").exists?

@@ -38,7 +38,7 @@ module Jobs
      return if locales.blank?

      locales.each do |locale|
-        next if DiscourseAi::Translation::LocaleNormalizer.is_same?(locale, detected_locale)
+        next if LocaleNormalizer.is_same?(locale, detected_locale)
        regionless_locale = locale.split("_").first
        next if topic.topic_localizations.where("locale LIKE ?", "#{regionless_locale}%").exists?

@@ -29,7 +29,7 @@ module Jobs
        missing_locales = locales - existing_locales - [category.locale]
        missing_locales.each do |locale|
          break if remaining_limit <= 0
-          next if DiscourseAi::Translation::LocaleNormalizer.is_same?(locale, category.locale)
+          next if LocaleNormalizer.is_same?(locale, category.locale)

          begin
            DiscourseAi::Translation::CategoryLocalizer.localize(category, locale)
@@ -12,6 +12,7 @@ class AiPersona < ActiveRecord::Base
  validates :system_prompt, presence: true, length: { maximum: 10_000_000 }
  validate :system_persona_unchangeable, on: :update, if: :system
  validate :chat_preconditions
+  validate :allowed_seeded_model, if: :default_llm_id
  validate :well_formated_examples
  validates :max_context_posts, numericality: { greater_than: 0 }, allow_nil: true
  # leaves some room for growth but sets a maximum to avoid memory issues
@@ -364,6 +365,17 @@ class AiPersona < ActiveRecord::Base
    end
  end

+  def allowed_seeded_model
+    return if default_llm_id.blank?
+
+    return if default_llm.nil?
+    return if !default_llm.seeded?
+
+    return if SiteSetting.ai_bot_allowed_seeded_models_map.include?(default_llm.id.to_s)
+
+    errors.add(:default_llm, I18n.t("discourse_ai.llm.configuration.invalid_seeded_model"))
+  end
+
  def well_formated_examples
    return if examples.blank?

@@ -99,7 +99,7 @@ class LlmModel < ActiveRecord::Base
  end

  def identifier
-    "#{id}"
+    "custom:#{id}"
  end

  def toggle_companion_user
@@ -1,46 +0,0 @@
-import Component from "@glimmer/component";
-import { tracked } from "@glimmer/tracking";
-import { ajax } from "discourse/lib/ajax";
-import { i18n } from "discourse-i18n";
-import SiteSettingComponent from "admin/components/site-setting";
-import SiteSetting from "admin/models/site-setting";
-
-export default class AiDefaultLlmSelector extends Component {
-  @tracked defaultLlmSetting = null;
-
-  constructor() {
-    super(...arguments);
-    this.#loadDefaultLlmSetting();
-  }
-
-  async #loadDefaultLlmSetting() {
-    const { site_settings } = await ajax("/admin/config/site_settings.json", {
-      data: {
-        plugin: "discourse-ai",
-        category: "discourse_ai",
-      },
-    });
-
-    const defaultLlmSetting = site_settings.find(
-      (setting) => setting.setting === "ai_default_llm_model"
-    );
-
-    this.defaultLlmSetting = SiteSetting.create(defaultLlmSetting);
-  }
-
-  <template>
-    <div class="ai-configure-default-llm">
-      <div class="ai-configure-default-llm__header">
-        <h3>{{i18n "discourse_ai.default_llm.title"}}</h3>
-        <p>{{i18n "discourse_ai.default_llm.description"}}</p>
-      </div>
-
-      {{#if this.defaultLlmSetting}}
-        <SiteSettingComponent
-          @setting={{this.defaultLlmSetting}}
-          class="ai-configure-default-llm__setting"
-        />
-      {{/if}}
-    </div>
-  </template>
-}
@@ -9,7 +9,6 @@ import DPageSubheader from "discourse/components/d-page-subheader";
 import DSelect from "discourse/components/d-select";
 import FilterInput from "discourse/components/filter-input";
 import { i18n } from "discourse-i18n";
-import AiDefaultLlmSelector from "./ai-default-llm-selector";
 import AiFeaturesList from "./ai-features-list";

 const ALL = "all";
@@ -203,8 +202,6 @@ export default class AiFeatures extends Component {
        />
      </div>

-      <AiDefaultLlmSelector />
-
      {{#if this.filteredFeatures.length}}
        <AiFeaturesList @modules={{this.filteredFeatures}} />
      {{else}}
@@ -9,7 +9,6 @@ import I18n, { i18n } from "discourse-i18n";
 import AdminSectionLandingItem from "admin/components/admin-section-landing-item";
 import AdminSectionLandingWrapper from "admin/components/admin-section-landing-wrapper";
 import DTooltip from "float-kit/components/d-tooltip";
-import AiDefaultLlmSelector from "./ai-default-llm-selector";
 import AiLlmEditor from "./ai-llm-editor";

 function isPreseeded(llm) {
@@ -138,9 +137,6 @@ export default class AiLlmsListEditor extends Component {
          }}
          @learnMoreUrl="https://meta.discourse.org/t/discourse-ai-large-language-model-llm-settings-page/319903"
        />
-
-        <AiDefaultLlmSelector />
-
        {{#if this.hasLlmElements}}
          <section class="ai-llms-list-editor__configured">
            <DPageSubheader
@@ -90,7 +90,7 @@ export default class AiSpam extends Component {
    this.isEnabled = model.is_enabled;

    if (model.llm_id) {
-      this.selectedLLM = model.llm_id;
+      this.selectedLLM = "custom:" + model.llm_id;
    } else {
      if (this.availableLLMs.length) {
        this.selectedLLM = this.availableLLMs[0].id;
@@ -127,7 +127,7 @@ export default class AiSpam extends Component {
  }

  get llmId() {
-    return this.selectedLLM;
+    return this.selectedLLM.toString().split(":")[1];
  }

  @action
@@ -158,51 +158,3 @@
 .ai-expanded-list__toggle-button {
  padding: 0;
 }
-
-.ai-configure-default-llm {
-  display: flex;
-  align-items: center;
-  gap: var(--space-2);
-  background: var(--primary-very-low);
-  padding: 1rem;
-  margin-block: 1rem;
-  border-radius: var(--d-border-radius);
-
-  &__header {
-    flex: 3;
-  }
-
-  &__setting {
-    flex: 2;
-    display: flex;
-    gap: var(--space-2);
-    align-items: center;
-    justify-content: center;
-
-    .setting-controls__undo {
-      display: none;
-    }
-  }
-
-  h3 {
-    color: var(--primary);
-    font-size: var(--font-0);
-    margin-bottom: 0;
-  }
-
-  p {
-    margin-top: 0;
-    font-size: var(--font-down-1);
-    color: var(--primary-high-or-secondary-low);
-  }
-
-  .select-kit,
-  .btn {
-    font-size: var(--font-down-1);
-  }
-
-  .setting-label,
-  .desc {
-    display: none;
-  }
-}
@@ -180,14 +180,13 @@ en:
            temperature:
              label: "Temperature"
              description: "Temperature to use for the LLM. Increase to increase randomness (leave empty to use model default)"
+            max_output_tokens:
+              label: "Max output tokens"
+              description: "When specified, sets an upper bound to the maximum number of tokens the model can generate. Respects LLM's max output tokens limit"

    discourse_ai:
      title: "AI"

-      default_llm:
-        title: "Default LLM model"
-        description: "The default LLM model to use for all AI features. This will be used if no LLM is specified in the feature configuration or persona. If no default LLM is specified, the last created LLM will be used."
-
      features:
        short_title: "Features"
        description: "These are the AI features available to visitors on your site. These can be configured to use specific personas and LLMs, and can be access controlled by groups."
@@ -570,7 +569,7 @@ en:
          max_usages_required: "Must be set if max tokens is not set"
        usage:
          ai_bot: "AI bot"
-          ai_helper: "Helper (%{persona})"
+          ai_helper: "Helper"
          ai_helper_image_caption: "Image caption"
          ai_persona: "Persona (%{persona})"
          ai_summarization: "Summarize"
@@ -25,7 +25,6 @@ en:
        description: "Periodic report based on a large language model"
  site_settings:
    discourse_ai_enabled: "Enable the discourse AI plugin."
-    ai_default_llm_model: "The default LLM model to use for all AI features"
    ai_artifact_security: "The AI artifact system generates IFRAMEs with runnable code. Strict mode forces an extra click to run code. Lax mode runs code immediately. Hybrid mode allows user to supply data-ai-artifact-autorun to show right away. Disabled mode disables the artifact system."
    ai_toxicity_enabled: "Enable the toxicity module."
    ai_toxicity_inference_service_api_endpoint: "URL where the API is running for the toxicity module"
@@ -583,7 +582,7 @@ en:
    llm:
      configuration:
        create_llm: "You need to setup an LLM before enabling this feature"
-        disable_modules_first: "You must disable these modules first: %{settings}"
+        disable_module_first: "You have to disable %{setting} first."
        set_llm_first: "Set %{setting} first"
        model_unreachable: "We couldn't get a response from this model. Check your settings first."
        invalid_seeded_model: "You can't use this model with this feature"
@@ -1,6 +1,6 @@
 discourse_ai:
  discourse_ai_enabled:
-    default: true
+    default: false
    client: true
  ai_artifact_security:
    client: true
@@ -11,12 +11,6 @@ discourse_ai:
      - "lax"
      - "hybrid"
      - "strict"
-  ai_default_llm_model:
-    default: ""
-    type: enum
-    allow_any: false
-    enum: "DiscourseAi::Configuration::LlmEnumerator"
-    validator: "DiscourseAi::Configuration::LlmValidator"

  ai_sentiment_enabled:
    default: false
@@ -109,7 +103,7 @@ discourse_ai:
    default: false
    client: true
    area: "ai-features/ai_helper"
-  ai_helper_model:  # Deprecated. TODO(keegan): Remove 2025-09-01
+  ai_helper_model:
    default: ""
    allow_any: false
    type: enum
@@ -155,7 +149,7 @@ discourse_ai:
      - "context_menu"
      - "image_caption"
    area: "ai-features/ai_helper"
-  ai_helper_image_caption_model:  # Deprecated. TODO(keegan): Remove 2025-09-01
+  ai_helper_image_caption_model:
    default: ""
    type: enum
    enum: "DiscourseAi::Configuration::LlmVisionEnumerator"
@@ -266,12 +260,12 @@ discourse_ai:
    client: true
    validator: "DiscourseAi::Configuration::LlmDependencyValidator"
    area: "ai-features/embeddings"
-  ai_embeddings_semantic_search_hyde_model:  # Deprecated. TODO(keegan): Remove 2025-09-01
-    hidden: true
+  ai_embeddings_semantic_search_hyde_model:
    default: ""
    type: enum
    allow_any: false
    enum: "DiscourseAi::Configuration::LlmEnumerator"
+    validator: "DiscourseAi::Configuration::LlmValidator"
    area: "ai-features/embeddings"
  ai_embeddings_semantic_search_hyde_model_allowed_seeded_models:
    default: ""
@@ -322,11 +316,12 @@ discourse_ai:
    client: true
    validator: "DiscourseAi::Configuration::LlmDependencyValidator"
    area: "ai-features/summarization"
-  ai_summarization_model: # Deprecated. TODO(keegan): Remove 2025-09-01
+  ai_summarization_model:
    default: ""
    allow_any: false
    type: enum
    enum: "DiscourseAi::Configuration::LlmEnumerator"
+    validator: "DiscourseAi::Configuration::LlmValidator"
    hidden: true
  ai_summarization_persona:
    default: "-11"
@@ -488,11 +483,12 @@ discourse_ai:
    client: true
    validator: "DiscourseAi::Configuration::LlmDependencyValidator"
    area: "ai-features/translation"
-  ai_translation_model: # Deprecated. TODO(keegan): Remove 2025-09-01
+  ai_translation_model:
    default: ""
    type: enum
    allow_any: false
    enum: "DiscourseAi::Configuration::LlmEnumerator"
+    validator: "DiscourseAi::Configuration::LlmValidator"
    area: "ai-features/translation"
  ai_translation_locale_detector_persona:
    default: "-27"
@@ -531,6 +527,8 @@ discourse_ai:
    area: "ai-features/translation"
  ai_translation_backfill_max_age_days:
    default: 5
+    min: 0
+    max: 20000
    client: false
    area: "ai-features/translation"
  ai_translation_verbose_logs:
@@ -1,16 +0,0 @@
-# frozen_string_literal: true
-class SeedAiDefaultLlmModel < ActiveRecord::Migration[7.2]
-  def up
-    return if DB.query_single("SELECT 1 FROM llm_models LIMIT 1").empty?
-
-    last_model_id = DB.query_single("SELECT id FROM llm_models ORDER BY id DESC LIMIT 1").first
-
-    if last_model_id.present?
-      execute "UPDATE site_settings SET value = '#{last_model_id}' WHERE name = 'ai_default_llm_model' AND (value IS NULL OR value = '');"
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -1,33 +0,0 @@
-# frozen_string_literal: true
-class CopyAiSummarizationModelToPersonaDefault < ActiveRecord::Migration[7.2]
-  def up
-    ai_summarization_model =
-      DB.query_single("SELECT value FROM site_settings WHERE name = 'ai_summarization_model'").first
-
-    if ai_summarization_model.present? && ai_summarization_model.start_with?("custom:")
-      # Extract the model ID from the setting value (e.g., "custom:-5" -> "-5")
-      model_id = ai_summarization_model.split(":").last
-
-      persona_settings = %w[ai_summarization_persona ai_summary_gists_persona]
-      default_persona_ids = [-11, -12]
-
-      persona_ids_query =
-        persona_settings
-          .map { |setting| "SELECT value FROM site_settings WHERE name = '#{setting}'" }
-          .join(" UNION ")
-      persona_ids = DB.query_single(persona_ids_query).compact
-      all_persona_ids = (default_persona_ids + persona_ids).map(&:to_i).uniq.join(",")
-
-      # Update the summarization personas with the extracted model ID
-      execute(<<~SQL)
-        UPDATE ai_personas
-        SET default_llm_id = #{model_id}
-        WHERE id IN (#{all_persona_ids}) AND default_llm_id IS NULL
-      SQL
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -1,43 +0,0 @@
-# frozen_string_literal: true
-class CopyAiHelperModelToPersonaDefault < ActiveRecord::Migration[7.2]
-  def up
-    ai_helper_model =
-      DB.query_single("SELECT value FROM site_settings WHERE name = 'ai_helper_model'").first
-
-    if ai_helper_model.present? && ai_helper_model.start_with?("custom:")
-      # Extract the model ID from the setting value (e.g., "custom:1" -> "1")
-      model_id = ai_helper_model.split(":").last
-
-      # Get persona IDs from site settings
-      persona_settings = %w[
-        ai_helper_proofreader_persona
-        ai_helper_title_suggestions_persona
-        ai_helper_explain_persona
-        ai_helper_smart_dates_persona
-        ai_helper_translator_persona
-        ai_helper_markdown_tables_persona
-        ai_helper_custom_prompt_persona
-      ]
-
-      default_persona_ids = [-18, -19, -20, -22, -23, -24, -25]
-      persona_ids_query =
-        persona_settings
-          .map { |setting| "SELECT value FROM site_settings WHERE name = '#{setting}'" }
-          .join(" UNION ")
-      persona_ids = DB.query_single(persona_ids_query).compact
-
-      all_persona_ids = (default_persona_ids + persona_ids).map(&:to_i).uniq.join(",")
-
-      # Update the helper personas with the extracted model ID
-      execute(<<~SQL)
-            UPDATE ai_personas
-            SET default_llm_id = #{model_id}
-            WHERE id IN (#{all_persona_ids}) AND default_llm_id IS NULL
-          SQL
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -1,35 +0,0 @@
-# frozen_string_literal: true
-class CopyHydeModelToPersona < ActiveRecord::Migration[7.2]
-  def up
-    hyde_model =
-      DB.query_single(
-        "SELECT value FROM site_settings WHERE name = 'ai_embeddings_semantic_search_hyde_model'",
-      ).first
-
-    if hyde_model.present? && hyde_model.start_with?("custom:")
-      # Extract the model ID from the setting value (e.g., "custom:1" -> "1")
-      model_id = hyde_model.split(":").last
-
-      persona_settings = %w[ai_embeddings_semantic_search_hyde_persona]
-      default_persona_ids = [-32]
-      persona_ids_query =
-        persona_settings
-          .map { |setting| "SELECT value FROM site_settings WHERE name = '#{setting}'" }
-          .join(" UNION ")
-      persona_ids = DB.query_single(persona_ids_query).compact
-
-      all_persona_ids = (default_persona_ids + persona_ids).map(&:to_i).uniq.join(",")
-
-      # Update the hyde persona with the extracted model ID
-      execute(<<~SQL)
-        UPDATE ai_personas
-        SET default_llm_id = #{model_id}
-        WHERE id IN (#{all_persona_ids}) AND default_llm_id IS NULL
-      SQL
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -1,39 +0,0 @@
-# frozen_string_literal: true
-class CopyTranslationModelToPersona < ActiveRecord::Migration[7.2]
-  def up
-    ai_translation_model =
-      DB.query_single("SELECT value FROM site_settings WHERE name = 'ai_translation_model'").first
-
-    if ai_translation_model.present? && ai_translation_model.start_with?("custom:")
-      # Extract the model ID from the setting value (e.g., "custom:-5" -> "-5")
-      model_id = ai_translation_model.split(":").last
-
-      persona_settings = %w[
-        ai_translation_locale_detector_persona
-        ai_translation_post_raw_translator_persona
-        ai_translation_topic_title_translator_persona
-        ai_translation_short_text_translator_persona
-      ]
-      default_persona_ids = [-27, -28, -29, -30]
-
-      persona_ids_query =
-        persona_settings
-          .map { |setting| "SELECT value FROM site_settings WHERE name = '#{setting}'" }
-          .join(" UNION ")
-      persona_ids = DB.query_single(persona_ids_query).compact
-
-      all_persona_ids = (default_persona_ids + persona_ids).map(&:to_i).uniq.join(",")
-
-      # Update the translation personas (IDs -27, -28, -29, -30) with the extracted model ID
-      execute(<<~SQL)
-        UPDATE ai_personas
-        SET default_llm_id = #{model_id}
-        WHERE id IN (#{all_persona_ids}) AND default_llm_id IS NULL
-      SQL
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -1,37 +0,0 @@
-# frozen_string_literal: true
-class CopyAiImageCaptionModelToPersonaDefault < ActiveRecord::Migration[7.2]
-  def up
-    ai_helper_image_caption_model =
-      DB.query_single(
-        "SELECT value FROM site_settings WHERE name = 'ai_helper_image_caption_model'",
-      ).first
-
-    if ai_helper_image_caption_model.present? &&
-         ai_helper_image_caption_model.start_with?("custom:")
-      # Extract the model ID from the setting value (e.g., "custom:1" -> "1")
-      model_id = ai_helper_image_caption_model.split(":").last
-
-      persona_settings = %w[ai_helper_post_illustrator_persona ai_helper_image_caption_persona]
-      default_persona_ids = [-21, -26]
-
-      persona_ids_query =
-        persona_settings
-          .map { |setting| "SELECT value FROM site_settings WHERE name = '#{setting}'" }
-          .join(" UNION ")
-      persona_ids = DB.query_single(persona_ids_query).compact
-
-      all_persona_ids = (default_persona_ids + persona_ids).map(&:to_i).uniq.join(",")
-
-      # Update the helper personas with the extracted model ID
-      execute(<<~SQL)
-        UPDATE ai_personas
-        SET default_llm_id = #{model_id}
-        WHERE id IN (#{all_persona_ids}) AND default_llm_id IS NULL
-      SQL
-    end
-  end
-
-  def down
-    raise ActiveRecord::IrreversibleMigration
-  end
-end
@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+class SetTranslationBackfillMaxAge < ActiveRecord::Migration[7.2]
+  def up
+    execute <<~SQL
+      UPDATE site_settings
+      SET value = '20000'
+      WHERE name = 'ai_translation_backfill_max_age_days'
+      AND value::integer > 20000;
+    SQL
+
+    execute <<~SQL
+      UPDATE site_settings
+      SET value = '0'
+      WHERE name = 'ai_translation_backfill_max_age_days'
+      AND value::integer < 0;
+    SQL
+  end
+
+  def down
+    raise ActiveRecord::IrreversibleMigration
+  end
+end
@@ -0,0 +1,23 @@
+# frozen_string_literal: true
+
+class EnableAiIfAlreadyInstalled < ActiveRecord::Migration[7.2]
+  def up
+    installed_at = DB.query_single(<<~SQL)&.first
+      SELECT created_at FROM schema_migration_details WHERE version='20230224165056'
+    SQL
+
+    if installed_at && installed_at < 1.hour.ago
+      # The plugin was installed before we changed it to be disabled-by-default
+      # Therefore, if there is no existing database value, enable the plugin
+      execute <<~SQL
+        INSERT INTO site_settings(name, data_type, value, created_at, updated_at)
+        VALUES('discourse_ai_enabled', 5, 't', NOW(), NOW())
+        ON CONFLICT (name) DO NOTHING
+      SQL
+    end
+  end
+
+  def down
+    raise ActiveRecord::IrreversibleMigration
+  end
+end
@@ -0,0 +1,11 @@
+# frozen_string_literal: true
+
+class AddIndexToAiTopicsEmbeddings < ActiveRecord::Migration[7.2]
+  def up
+    add_index :ai_topics_embeddings, %i[topic_id model_id]
+  end
+
+  def down
+    raise ActiveRecord::IrreversibleMigration
+  end
+end
@@ -24,6 +24,7 @@ if defined?(DiscourseAutomation)
    field :max_post_tokens, component: :text
    field :stop_sequences, component: :text_list, required: false
    field :temperature, component: :text
+    field :max_output_tokens, component: :text

    # Actions
    field :category, component: :category
@@ -85,6 +86,9 @@ if defined?(DiscourseAutomation)
        temperature = temperature.to_f
      end

+      max_output_tokens = fields.dig("max_output_tokens", "value").to_i
+      max_output_tokens = nil if max_output_tokens <= 0
+
      max_post_tokens = nil if max_post_tokens <= 0

      stop_sequences = fields.dig("stop_sequences", "value")
@@ -122,6 +126,7 @@ if defined?(DiscourseAutomation)
          stop_sequences: stop_sequences,
          automation: self.automation,
          temperature: temperature,
+          max_output_tokens: max_output_tokens,
          action: context["action"],
        )
      rescue => e
@@ -312,9 +312,18 @@ module DiscourseAi

      # Priorities are:
      #   1. Persona's default LLM
-      #   2. SiteSetting.ai_default_llm_id (or newest LLM if not set)
+      #   2. Hidden `ai_helper_model` setting, or `ai_helper_image_caption_model` for image_caption.
+      #   3. Newest LLM config
      def self.find_ai_helper_model(helper_mode, persona_klass)
-        model_id = persona_klass.default_llm_id || SiteSetting.ai_default_llm_model
+        model_id = persona_klass.default_llm_id
+
+        if !model_id
+          if helper_mode == IMAGE_CAPTION
+            model_id = SiteSetting.ai_helper_image_caption_model&.split(":")&.last
+          else
+            model_id = SiteSetting.ai_helper_model&.split(":")&.last
+          end
+        end

        if model_id.present?
          LlmModel.find_by(id: model_id)
@@ -12,14 +12,10 @@ module DiscourseAi
        return nil if content.blank?

        suggested_title = call_llm(content)
-        return nil if suggested_title.blank?
-        
        cleanup(suggested_title)
      end

      def call_llm(thread_content)
-        return nil if SiteSetting.ai_default_llm_model.blank?
-        
        chat = "<input>\n#{thread_content}\n</input>"

        prompt =
@@ -34,7 +30,7 @@ module DiscourseAi
            messages: [{ type: :user, content: chat, id: "User" }],
          )

-        DiscourseAi::Completions::Llm.proxy(SiteSetting.ai_default_llm_model).generate(
+        DiscourseAi::Completions::Llm.proxy(SiteSetting.ai_helper_model).generate(
          prompt,
          user: Discourse.system_user,
          stop_sequences: ["</input>"],
@@ -57,12 +57,6 @@ module DiscourseAi
      end

      def diffusion_prompt(text, user)
-        llm_model =
-          AiPersona.find_by(id: SiteSetting.ai_helper_post_illustrator_persona)&.default_llm_id ||
-            SiteSetting.ai_default_llm_model
-
-        return nil if llm_model.blank?
-
        prompt =
          DiscourseAi::Completions::Prompt.new(
            <<~TEXT.strip,
@@ -72,7 +66,7 @@ module DiscourseAi
            messages: [{ type: :user, content: text, id: user.username }],
          )

-        DiscourseAi::Completions::Llm.proxy(llm_model).generate(
+        DiscourseAi::Completions::Llm.proxy(SiteSetting.ai_helper_model).generate(
          prompt,
          user: user,
          feature_name: "illustrate_post",
@@ -37,7 +37,7 @@ module DiscourseAi
            value_h["id"] > 0 ||
              SiteSetting.ai_automation_allowed_seeded_models_map.include?(value_h["id"].to_s)
          end
-          .each { |value_h| value_h["id"] = "#{value_h["id"]}" }
+          .each { |value_h| value_h["id"] = "custom:#{value_h["id"]}" }

      values
    end
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
dependabot[bot]	415c1156e0	Build(deps-dev): Bump rack from 3.1.16 to 3.1.18 Bumps [rack](https://github.com/rack/rack) from 3.1.16 to 3.1.18. - [Release notes](https://github.com/rack/rack/releases) - [Changelog](https://github.com/rack/rack/blob/main/CHANGELOG.md) - [Commits](https://github.com/rack/rack/compare/v3.1.16...v3.1.18) --- updated-dependencies: - dependency-name: rack dependency-version: 3.1.18 dependency-type: indirect ... Signed-off-by: dependabot[bot] <[email protected]>	2025-12-27 22:27:24 +00:00
Jarek Radosz	c312b3184a	DEV: Mention the core move in readme (#1514 )	2025-07-22 11:08:32 +01:00
Jarek Radosz	f231aad8b5	DEV: Disable the plugin by default (#1511 ) …and preserve the current setting on existing sites	2025-07-22 11:05:52 +01:00
Alan Guo Xiang Tan	cc77e73cfd	PERF: Create index `ON ai_topics_embeddings(topic_id, model_id)` (#1513 ) This index helps to speed up queries that joins the `topics` table against the `ai_topics_embeddings` table on the `topic_id` column. There are a number of queries which filters on `ai_topics_embeddings.model_id` so we are including that in the index as well.	2025-07-22 16:46:56 +08:00
Natalie Tay	1d88ca44f0	DEV: Use core's locale normalizer (#1509 ) We've moved the class to core here discourse/discourse#33702 so following up a deletion in this repo.	2025-07-22 10:09:03 +08:00
Roman Rizzi	6059b6e111	FIX: Customizable max_output_tokens for AI triage. (#1510 ) We enforced a hard limit of 700 tokens in this script, which is not enough when using thinking models, which can quickly use all of them. A temporary solution could be bumping the limit, but there is no guarantee we won't hit it again, and it's hard to find one value that fits all scenarios. Another alternative could be removing it and relying on the LLM config's `max_output_token`, but if you want different rules and want to assign different limits, you are forced to duplicate the config each time. Considering all this, we are adding a dedicated field for this in the triage script, giving you an easy way to tweak it to your needs. If empty, no limit is applied.	2025-07-21 15:36:39 -03:00
Natalie Tay	5d80a34589	FIX: Add a max token limit based on the text to be translated (#1507 ) We're seeing that some LLMs are using 65000+ tokens for raw text that is only 10-1000 characters long. This PR adds a max_token to be passed to the LLM API for each translation based on the length of the text.	2025-07-17 17:47:15 +08:00
Natalie Tay	8630bc145e	DEV: Set a min and max for translations backfill (#1508 ) Since we use the value for days.ago and are seeing PG::DatetimeFieldOverflow: ERROR: timestamp out of range: "5473790-07-13 08:43:28.497823 BC, set limits to the site setting.	2025-07-17 17:43:05 +08:00