Skip to content

feat(component): add Ollama LLM and Embedding providers - #365

Open
Timelovers wants to merge 1 commit into
EverMind-AI:mainfrom
Timelovers:feat/ollama-providers
Open

feat(component): add Ollama LLM and Embedding providers#365
Timelovers wants to merge 1 commit into
EverMind-AI:mainfrom
Timelovers:feat/ollama-providers

Conversation

@Timelovers

@Timelovers Timelovers commented Jul 27, 2026

Copy link
Copy Markdown

Description

Add OllamaProvider (LLM) and OllamaEmbeddingProvider (embedding) — thin subclasses of the existing OpenAI-compatible providers with Ollama-specific defaults.

Ollama exposes an OpenAI-compatible API at http://localhost:11434/v1 with api-key: ollama. These providers extend the existing implementations so users get sensible local-first defaults out of the box — no need to remember the base URL or supply a dummy key.

from everos.component.llm import OllamaProvider
llm = OllamaProvider(model="llama3.1")

from everos.component.embedding import OllamaEmbeddingProvider
embed = OllamaEmbeddingProvider(model="nomic-embed-text")

The existing OpenAIProvider / OpenAIEmbeddingProvider already work with Ollama via .env config — this PR just makes the setup explicit and discoverable with zero new logic.

Type of Change

  • New feature (non-breaking change that adds functionality)

How Has This Been Tested?

  • Unit Test — 8 tests covering defaults, custom params, inheritance, and public imports
  • Syntax check — all files pass ast.parse()

Checklist

  • I have performed a self-review of my own code
  • I have commented my code in hard-to-understand areas
  • I have added tests that prove my fix is effective or that my feature works
  • I have created related documentation issue/PR (if applicable)
  • New and existing unit tests pass locally with make test

Related Issues

N/A — incremental addition following the existing one-provider-per-file pattern.
@Timelovers
feat(component): add Ollama LLM and Embedding providers
6260a02 
Add OllamaProvider and OllamaEmbeddingProvider — thin subclasses of
OpenAIProvider / OpenAIEmbeddingProvider with Ollama-specific defaults
(base_url=http://localhost:11434/v1, api_key='ollama', dim=768).

Ollama's API is OpenAI-compatible, so these providers extend the
existing implementations with no new logic — just sensible local-first
defaults. Users can now do:

    from everos.component.llm import OllamaProvider
    llm = OllamaProvider(model='llama3.1')

instead of remembering base_url and supplying a dummy key.

Also add 8 unit tests (defaults, custom params, inheritance, imports).

Co-authored-by: Timelovers <46080686+Timelovers@users.noreply.github.com>
@Timelovers

Timelovers commented Jul 28, 2026

Copy link
Copy Markdown
Author

Hi — just checking if this is on the radar. Happy to adjust anything if needed. Thanks!
@Timelovers

Timelovers commented Aug 2, 2026

Copy link
Copy Markdown
Author

Hi @cyfyifanchen — friendly ping on this PR. It adds Ollama providers for fully local deployment. Small change, no logic modifications. Would appreciate a review when you have a moment. Thanks!

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

1 participant

@Timelovers