lex-llm-ollama

LegionIO LLM provider extension for Ollama.

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.

Load it with require 'legion/extensions/llm/ollama'.

What It Provides

• Ollama-native chat requests through POST /api/chat
• Streaming chat support
• Model discovery through GET /api/tags with automatic embedding capability inference
• Running model inspection through GET /api/ps
• Model details through POST /api/show
• 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
• 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)
└── (shared from lex-llm)
    ├── RegistryPublisher      # Best-effort async llm.registry event publishing
    ├── RegistryEventBuilder   # Sanitized registry envelope construction
    └── Transport/             # Shared exchange and message classes

Defaults

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: {}
# }

Configuration

Legion::Extensions::Llm.configure do |config|
  config.default_model = "qwen3.6:27b"
  config.default_embedding_model = "nomic-embed-text:latest"
end

Development

bundle install
bundle exec rspec       # 0 failures
bundle exec rubocop -A  # auto-fix
bundle exec rubocop     # lint check

License

Apache-2.0