Implement Ollama Cloud provider#63
Conversation
Add replace directives for local fantasy and catwalk repos to test Ollama Cloud integration from PRs charmbracelet/fantasy#63 and charmbracelet/catwalk#96.
|
@vorticalbox is it possible to squash/rebase and cleanup the history a bit, it looks like there are a lot of commits that should not be related to this PR. In addition can you add some tests under the |
|
I will find some time today to sort this out.
Sent from [Proton Mail](https://proton.me/mail/home) for iOS.
…-------- Original Message --------
On Monday, 01/05/26 at 10:26 Kujtim Hoxha ***@***.***> wrote:
kujtimiihoxha left a comment [(charmbracelet/fantasy#63)](#63 (comment))
***@***.***(https://github.com/vorticalbox) is it possible to squash/rebase and cleanup the history a bit, it looks like there are a lot of commits that should not be related to this PR.
In addition can you add some tests under the providertests so we can be sure this is working? you can use one of the other providers as an example.
—
Reply to this email directly, [view it on GitHub](#63 (comment)), or [unsubscribe](https://github.com/notifications/unsubscribe-auth/ACTBXMN72YAXA6J2D5CT56L4FI345AVCNFSM6AAAAACLKT6MR6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOMBZHAZTMOBVGI).
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@vorticalbox I can see that you added a test, for the tests to work in the CI (and for other users when testing) you need to run the tests once locally with a real API key and that will record the requests/results (this will not record the API key). |
There was a problem hiding this comment.
Pull request overview
This PR implements a new Ollama Cloud provider for the fantasy AI SDK, enabling integration with Ollama's cloud-based language models.
- Adds complete provider implementation with support for text generation, streaming, tool calls, and reasoning
- Implements provider options for configuring the "think" feature
- Includes test suite following the established testing patterns
Reviewed changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
providers/ollamacloud/ollamacloud.go |
Core provider implementation with HTTP client setup, authentication, and request handling |
providers/ollamacloud/language_model.go |
Language model implementation with Generate/Stream methods, tool call support, and reasoning capabilities |
providers/ollamacloud/provider_options.go |
Provider-specific options including the "think" parameter for reasoning |
providertests/ollamacloud_test.go |
Test suite with common tests for two Ollama Cloud models |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Implement Ollama Cloud provider with native API support: - Custom HTTP client for /api/chat endpoint - Support for reasoning (thinking) tokens - Tool calling support - Streaming and non-streaming generation - Provider-specific options (think parameter) - Basic provider tests The provider uses Ollama's native API format which differs from OpenAI's API, requiring a custom implementation.
CONTRIBUTING.md.