lex-llm-ollama 0.2.0 → 0.2.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 599e2c13614a5f4a912562dda8336e43cee73a4934d95a782e6d231374ad322e
-  data.tar.gz: aab71c8eb82f54d91da01f4a68547cfcdc33ac53cad06c02fad02ef0d901931c
+  metadata.gz: b4d5bd8102f3cc484e81ec31067f899306246ee6b333a87528874a6afedaffdc
+  data.tar.gz: 7a56c5e3608f88cc40075ae0ba249e05af48d8dc2487ec5ba4b300c054d7d65f
 SHA512:
-  metadata.gz: 33e4999dbe5781060c429020fccf0cbb0a090ef0cba142c0a136be53e81587a0f66efd053f3e8daa887134718b6a9429ff2bc8febad1cd0e307faf92c193612d
-  data.tar.gz: 6a43fb12fa810b332586e0bb92ec1ab600d82b089d2cb14db65b922708e3c590fabaaf724170ba5c04aa9f5367bde7cb05285ff31ef52e4b43875948fa94beda
+  metadata.gz: 375772149e7b1641cfea2f3446c08d2d50155f07bb92f3b90e7e7f00fac91541754e9f085531ea20ab7af7c49a6fdbb68145d9e9f5131d236493882e358e0420
+  data.tar.gz: f450168705b4b7b0dd340f9cdeed6294dc39bd7053902c23d8f820ea8310f3088206435cecd4bfa208d4f9aeba220127568cf379731f325a9d51042c802850c6

data/.github/workflows/ci.yml

@@ -8,8 +8,20 @@ jobs:
   ci:
     uses: LegionIO/.github/.github/workflows/ci.yml@main
 
+  excluded-files:
+    uses: LegionIO/.github/.github/workflows/excluded-files.yml@main
+
+  security:
+    uses: LegionIO/.github/.github/workflows/security-scan.yml@main
+
+  version-changelog:
+    uses: LegionIO/.github/.github/workflows/version-changelog.yml@main
+
+  dependency-review:
+    uses: LegionIO/.github/.github/workflows/dependency-review.yml@main
+
   release:
-    needs: ci
+    needs: [ci, excluded-files, security]
     if: github.event_name == 'push' && github.ref == 'refs/heads/main'
     uses: LegionIO/.github/.github/workflows/release.yml@main
     secrets:

data/CHANGELOG.md

@@ -1,5 +1,40 @@
 # Changelog
 
+## 0.2.6 - 2026-05-06
+
+- Load provider-owned fleet actors through the LegionIO subscription base and the canonical Ollama provider root.
+- Keep fleet runners anchored on the provider root namespace so provider constants and instance discovery are always loaded.
+- Preserve configured transport and tier metadata when Ollama builds routing offerings.
+- Gate release publishing on the shared security workflow.
+
+## 0.2.5 - 2026-05-06
+
+- Mark cached offering discovery fallback exceptions as handled.
+- Refresh README provider contract, fleet responder, development gate, and license details.
+
+## 0.2.4 - 2026-05-06
+
+- Use the shared `lex-llm` fleet provider responder helper for provider-owned fleet workers.
+- Remove the runtime `legion-llm` dependency and require `lex-llm >= 0.4.3` for responder-side fleet execution.
+
+## 0.2.3 - 2026-05-06
+
+- Remove require-time provider self-registration; `legion-llm` now owns adapter creation and registry writes from loaded provider discovery metadata.
+- Bump dependency floors to `lex-llm >= 0.4.1` and `legion-llm >= 0.9.1`.
+
+## 0.2.2 - 2026-05-06
+
+- Add provider contract specs for the shared keyword-only `lex-llm` provider API.
+- Move Ollama defaults back to `Legion::Extensions::Llm.provider_settings` with instance-level fleet responder settings.
+- Serve non-live Ollama offering reads from cached live model discovery instead of probing the configured endpoint.
+- Add provider-owned fleet responder actor and runner backed by `legion-llm` fleet policy execution.
+- Bump the transport dependency floor to `legion-transport >= 1.4.14`.
+
+## 0.2.1 - 2026-05-03
+
+- Normalize configured Ollama instance endpoint aliases to `base_url`.
+- Use instance `base_url` config before provider defaults.
+
 ## 0.2.0 - 2026-05-01
 
 - Add auto-discovery via CredentialSources and AutoRegistration from lex-llm 0.3.0

data/Gemfile

@@ -4,6 +4,8 @@ source 'https://rubygems.org'
 
 group :test do
   llm_base_path = ENV.fetch('LEX_LLM_PATH', File.expand_path('../lex-llm', __dir__))
+  transport_path = ENV.fetch('LEGION_TRANSPORT_PATH', File.expand_path('../../legion-transport', __dir__))
+  gem 'legion-transport', path: transport_path if File.directory?(transport_path)
   gem 'lex-llm', path: llm_base_path if File.directory?(llm_base_path)
 end
 

data/README.md

@@ -2,7 +2,7 @@
 
 LegionIO LLM provider extension for [Ollama](https://ollama.ai).
 
-This gem lives under `Legion::Extensions::Llm::Ollama` and depends on `lex-llm` (>= 0.1.9) for shared provider-neutral routing, fleet, transport, and registry primitives.
+This gem lives under `Legion::Extensions::Llm::Ollama` and depends on `lex-llm >= 0.4.3` for shared provider-neutral routing, response normalization, fleet envelopes, responder execution, transport, and registry primitives. It does not carry a runtime `legion-llm` dependency; `legion-llm` owns higher-level routing and can discover this provider through normal extension loading.
 
 Load it with `require 'legion/extensions/llm/ollama'`.
 
@@ -16,18 +16,22 @@ Load it with `require 'legion/extensions/llm/ollama'`.
 - Model download helper through `POST /api/pull`
 - Embeddings through `POST /api/embed`
 - Best-effort `llm.registry` availability events via the shared `Legion::Extensions::Llm::RegistryPublisher`
-- Full settings schema with model whitelist/blacklist, TLS, and multi-host base URL resolution
+- Local socket discovery plus configured instance discovery through the shared `lex-llm` credential sources
+- Provider-owned fleet response handling through `Legion::Extensions::Llm::Fleet::ProviderResponder`
 - Full `Legion::Logging::Helper` integration with structured `handle_exception` in every rescue block
 
 ## Architecture
 
 ```
 Legion::Extensions::Llm::Ollama
-├── Provider               # Ollama provider (chat, stream, embed, models, readiness)
+├── Provider                   # Ollama provider (chat, stream, embed, models, readiness)
+├── Actor::FleetWorker         # Optional provider-owned fleet subscription actor
+├── Runners::FleetWorker       # Delegates fleet execution to lex-llm
 └── (shared from lex-llm)
-    ├── RegistryPublisher      # Best-effort async llm.registry event publishing
-    ├── RegistryEventBuilder   # Sanitized registry envelope construction
-    └── Transport/             # Shared exchange and message classes
+    ├── Fleet::ProviderResponder
+    ├── RegistryPublisher
+    ├── RegistryEventBuilder
+    └── Transport/
 ```
 
 ## Defaults
@@ -35,35 +39,71 @@ Legion::Extensions::Llm::Ollama
 ```ruby
 Legion::Extensions::Llm::Ollama.default_settings
 # {
-#   enabled: false,
-#   base_url: '127.0.0.1:11434',
-#   default_model: 'qwen3.5:latest',
-#   model_whitelist: [],
-#   model_blacklist: [],
-#   model_cache_ttl: 60,
-#   tls: { enabled: false, verify: :peer },
-#   instances: {}
+#   enabled: true,
+#   provider_family: :ollama,
+#   instances: {
+#     default: {
+#       endpoint: 'http://127.0.0.1:11434',
+#       default_model: 'qwen3.5:latest',
+#       tier: :local,
+#       transport: :http,
+#       credentials: {},
+#       usage: { inference: true, embedding: true, image: false },
+#       limits: { concurrency: 1 },
+#       fleet: {
+#         enabled: false,
+#         respond_to_requests: false,
+#         capabilities: %i[chat stream_chat embed],
+#         lanes: [],
+#         concurrency: 1,
+#         queue_suffix: nil
+#       }
+#     }
+#   }
 # }
 ```
 
 ## Configuration
 
-```ruby
-Legion::Extensions::Llm.configure do |config|
-  config.default_model = "qwen3.6:27b"
-  config.default_embedding_model = "nomic-embed-text:latest"
-end
+`discover_instances` returns a local `http://127.0.0.1:11434` instance when the Ollama socket is reachable. Additional instances can be supplied under the shared LLM extension configuration and may use `base_url`, `endpoint`, `api_base`, or `ollama_api_base`; the extension normalizes those aliases to `base_url`.
+
+```yaml
+extensions:
+  llm:
+    ollama:
+      instances:
+        lab:
+          base_url: http://ollama-lab:11434
+          default_model: qwen3.5:latest
+```
+
+## Fleet Responder
+
+Provider instances can opt in to consuming Legion LLM fleet requests. The provider-owned fleet actor only starts when at least one discovered instance enables `respond_to_requests`, and the runner delegates execution to the shared `lex-llm` responder helper.
+
+```yaml
+extensions:
+  llm:
+    ollama:
+      instances:
+        local:
+          fleet:
+            enabled: true
+            respond_to_requests: true
+            capabilities:
+              - chat
+              - stream_chat
+              - embed
 ```
 
 ## Development
 
 ```bash
 bundle install
-bundle exec rspec       # 0 failures
-bundle exec rubocop -A  # auto-fix
-bundle exec rubocop     # lint check
+bundle exec rspec --format json --out tmp/rspec_results.json --format progress --out tmp/rspec_progress.txt
+bundle exec rubocop -A
 ```
 
 ## License
 
-Apache-2.0
+MIT

data/lex-llm-ollama.gemspec

@@ -26,5 +26,6 @@ Gem::Specification.new do |spec|
   spec.add_dependency 'legion-json', '>= 1.2.1'
   spec.add_dependency 'legion-logging', '>= 1.3.2'
   spec.add_dependency 'legion-settings', '>= 1.3.14'
-  spec.add_dependency 'lex-llm', '>= 0.3.0'
+  spec.add_dependency 'legion-transport', '>= 1.4.14'
+  spec.add_dependency 'lex-llm', '>= 0.4.3'
 end

data/lib/legion/extensions/llm/ollama/actors/fleet_worker.rb

@@ -0,0 +1,43 @@
+# frozen_string_literal: true
+
+begin
+  require 'legion/extensions/actors/subscription'
+rescue LoadError => e
+  warn(e.message) if $VERBOSE
+end
+
+unless defined?(Legion::Extensions::Actors::Subscription)
+  raise LoadError, 'LegionIO actor runtime is required for Ollama fleet worker'
+end
+
+require 'legion/extensions/llm/ollama'
+require 'legion/extensions/llm/fleet/provider_responder'
+
+module Legion
+  module Extensions
+    module Llm
+      module Ollama
+        module Actor
+          # Subscription actor for Ollama fleet request consumption.
+          class FleetWorker < Legion::Extensions::Actors::Subscription
+            def runner_class
+              'Legion::Extensions::Llm::Ollama::Runners::FleetWorker'
+            end
+
+            def runner_function
+              'handle_fleet_request'
+            end
+
+            def use_runner?
+              false
+            end
+
+            def enabled?
+              Legion::Extensions::Llm::Fleet::ProviderResponder.enabled_for?(Ollama.discover_instances)
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/legion/extensions/llm/ollama/provider.rb

@@ -41,7 +41,7 @@ module Legion
           end
 
           def config_base_url
-            settings[:base_url]
+            config.respond_to?(:base_url) ? config.base_url : settings[:base_url]
           end
 
           def completion_url = '/api/chat'
@@ -92,8 +92,62 @@ module Legion
             raise
           end
 
+          def discover_offerings(live: false, **)
+            models = if live
+                       @cached_models = list_models
+                     else
+                       Array(@cached_models)
+                     end
+            models.map { |model_info| offering_from_model(model_info) }
+          rescue StandardError => e
+            handle_exception(e, level: :warn, handled: true, operation: 'ollama.discover_offerings')
+            []
+          end
+
           private
 
+          def offering_from_model(model_info)
+            Legion::Extensions::Llm::Routing::ModelOffering.new(
+              provider_family: :ollama,
+              instance_id: config.respond_to?(:instance_id) ? config.instance_id : :default,
+              transport: offering_transport,
+              tier: offering_tier,
+              model: model_info.id,
+              usage_type: offering_usage_type(model_info),
+              capabilities: offering_capabilities(model_info),
+              limits: offering_limits(model_info),
+              metadata: offering_metadata(model_info)
+            )
+          end
+
+          def offering_transport
+            config.respond_to?(:transport) ? config.transport : :http
+          end
+
+          def offering_tier
+            config.respond_to?(:tier) ? config.tier : :local
+          end
+
+          def offering_usage_type(model_info)
+            model_info.embedding? ? :embedding : :inference
+          end
+
+          def offering_capabilities(model_info)
+            model_info.capabilities.map(&:to_s)
+          end
+
+          def offering_limits(model_info)
+            { context_window: model_info.context_length }.compact
+          end
+
+          def offering_metadata(model_info)
+            {
+              context_length: model_info.context_length,
+              family: model_info.family,
+              size_bytes: model_info.size_bytes
+            }.compact
+          end
+
           def ollama_keep_alive
             settings[:keep_alive]
           end

data/lib/legion/extensions/llm/ollama/runners/fleet_worker.rb

@@ -0,0 +1,30 @@
+# frozen_string_literal: true
+
+require 'legion/extensions/llm/fleet/provider_responder'
+require 'legion/extensions/llm/ollama'
+
+module Legion
+  module Extensions
+    module Llm
+      module Ollama
+        module Runners
+          # Runner entrypoint for Ollama fleet request execution.
+          module FleetWorker
+            module_function
+
+            def handle_fleet_request(payload, delivery: nil, properties: nil)
+              Legion::Extensions::Llm::Fleet::ProviderResponder.call(
+                payload: payload,
+                provider_family: Ollama::PROVIDER_FAMILY,
+                provider_class: Ollama::Provider,
+                provider_instances: -> { Ollama.discover_instances },
+                delivery: delivery,
+                properties: properties
+              )
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/legion/extensions/llm/ollama/version.rb

@@ -4,7 +4,7 @@ module Legion
   module Extensions
     module Llm
       module Ollama
-        VERSION = '0.2.0'
+        VERSION = '0.2.6'
       end
     end
   end

data/lib/legion/extensions/llm/ollama.rb

@@ -16,16 +16,26 @@ module Legion
         PROVIDER_FAMILY = :ollama
 
         def self.default_settings
-          {
-            enabled: false,
-            base_url: '127.0.0.1:11434',
-            default_model: 'qwen3.5:latest',
-            model_whitelist: [],
-            model_blacklist: [],
-            model_cache_ttl: 60,
-            tls: { enabled: false, verify: :peer },
-            instances: {}
-          }
+          ::Legion::Extensions::Llm.provider_settings(
+            family: PROVIDER_FAMILY,
+            instance: {
+              endpoint: 'http://127.0.0.1:11434',
+              default_model: 'qwen3.5:latest',
+              tier: :local,
+              transport: :http,
+              credentials: {},
+              usage: { inference: true, embedding: true, image: false },
+              limits: { concurrency: 1 },
+              fleet: {
+                enabled: false,
+                respond_to_requests: false,
+                capabilities: %i[chat stream_chat embed],
+                lanes: [],
+                concurrency: 1,
+                queue_suffix: nil
+              }
+            }
+          )
         end
 
         def self.provider_class
@@ -50,7 +60,7 @@ module Legion
           configured = CredentialSources.setting(:extensions, :llm, :ollama, :instances)
           if configured.is_a?(Hash)
             configured.each do |name, config|
-              instances[name.to_sym] = config.merge(
+              instances[name.to_sym] = normalize_instance_config(config).merge(
                 tier: :direct,
                 capabilities: %i[completion embedding vision]
               )
@@ -59,9 +69,15 @@ module Legion
 
           instances
         end
+
+        def self.normalize_instance_config(config)
+          normalized = config.to_h.transform_keys { |key| key.respond_to?(:to_sym) ? key.to_sym : key }
+          normalized[:base_url] ||= normalized.delete(:ollama_api_base)
+          normalized[:base_url] ||= normalized.delete(:api_base)
+          normalized[:base_url] ||= normalized.delete(:endpoint)
+          normalized.compact
+        end
       end
     end
   end
 end
-
-Legion::Extensions::Llm::Ollama.register_discovered_instances

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: lex-llm-ollama
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.2.6
 platform: ruby
 authors:
 - LegionIO
@@ -51,20 +51,34 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 1.3.14
+- !ruby/object:Gem::Dependency
+  name: legion-transport
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.4.14
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.4.14
 - !ruby/object:Gem::Dependency
   name: lex-llm
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.3.0
+        version: 0.4.3
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.3.0
+        version: 0.4.3
 description: Ollama provider integration for the LegionIO LLM routing framework.
 email:
 - matthewdiverson@gmail.com
@@ -83,7 +97,9 @@ files:
 - README.md
 - lex-llm-ollama.gemspec
 - lib/legion/extensions/llm/ollama.rb
+- lib/legion/extensions/llm/ollama/actors/fleet_worker.rb
 - lib/legion/extensions/llm/ollama/provider.rb
+- lib/legion/extensions/llm/ollama/runners/fleet_worker.rb
 - lib/legion/extensions/llm/ollama/version.rb
 homepage: https://github.com/LegionIO/lex-llm-ollama
 licenses: