fix(ai): update model list and opencode-go DeepSeek V4 Flash thinking-level expectation - #1218
Closed
snimu wants to merge 2 commits into
Closed
fix(ai): update model list and opencode-go DeepSeek V4 Flash thinking-level expectation#1218snimu wants to merge 2 commits into
snimu wants to merge 2 commits into
Conversation
The opencode-go zen catalog now serves deepseek-v4-flash through anthropic-messages instead of the DeepSeek openai-completions format, so the DeepSeek-specific high/xhigh thinking-level mapping no longer applies on that provider. packages/ai regenerates models.generated.ts from the live catalog on every build, so CI picked this up on every branch. The test now pins the anthropic-messages route and its default thinking levels, keeping the DeepSeek-provider and OpenRouter expectations unchanged.
…alogs Commits the current output of `npm run generate-models` so the checked-in catalog matches upstream again (opencode-go moved deepseek-v4-flash to anthropic-messages, plus assorted drift accumulated since the last regeneration).
snimu
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
The
Test (ai)CI job is failing on every branch, including main. Here's why: the build regenerates the model catalog (models.generated.ts) by fetching the providers' live model lists. One provider (opencode-go) changed how it serves DeepSeek V4 Flash — different API, different thinking levels. One of our tests still expects the old behavior, so it fails everywhere, no matter what the branch changes.Committing an updated catalog alone wouldn't help: the build fetches fresh data anyway and would overwrite it. The test itself has to change.
Fix
Two commits, nothing else:
npm run generate-modelsoutput, so the file in the repo matches reality again.What doesn't change
No code changes — one test file and one generated file.
Validation
packages/aitest suite with a fresh catalog: 315 passed, 0 failed