LearnNewsExamplesServices
Frontmatter
id9644
titleAdd Neo.ai.provider.Ollama base adapter
stateClosed
labels
aiarchitecturefeature
assigneestobiu
createdAtApr 3, 2026, 12:47 PM
updatedAtApr 3, 2026, 12:53 PM
githubUrlhttps://github.com/neomjs/neo/issues/9644
authortobiu
commentsCount1
parentIssue9639
subIssues[]
subIssuesCompleted0
subIssuesTotal0
blockedBy[]
blocking[]
closedAtApr 3, 2026, 12:53 PM

Add Neo.ai.provider.Ollama base adapter

Closedaiarchitecturefeature
tobiu
tobiu commented on Apr 3, 2026, 12:47 PM

Parent Epic: #9639

Problem

The framework needs to interface directly with a local ollama run gemma-4 daemon to handle the "Night Shift" REM mode tasks.

Solution

Create src/ai/provider/Ollama.mjs (or equivalent location) extending a base provider.

  • Implement generateContent mapped to the Ollama /api/generate endpoint.
  • Implement chat / sendMessage mapped to Ollama /api/chat.
  • Support system instructions and JSON object extraction mirroring the Gemini provider.
tobiu added the ai label on Apr 3, 2026, 12:47 PM
tobiu added the architecture label on Apr 3, 2026, 12:47 PM
tobiu added the feature label on Apr 3, 2026, 12:47 PM
tobiu added parent issue #9639 on Apr 3, 2026, 12:47 PM
tobiu
tobiu Apr 3, 2026, 12:50 PM

Implementation Complete: The Neo.ai.provider.Ollama adapter has been implemented.

tobiu assigned to @tobiu on Apr 3, 2026, 12:53 PM
tobiu closed this issue on Apr 3, 2026, 12:53 PM
tobiu referenced in commit ba22ba4 - "feat: Add Neo.ai.provider.Ollama base adapter (#9644)" on Apr 3, 2026, 12:53 PM