Repository navigation
feat(models): reachability-aware model picker (#1979) - #1992
Closed
aarontrowbridge wants to merge 14 commits into
Closed
aarontrowbridge wants to merge 14 commits into
aarontrowbridge wants to merge 14 commits into
Conversation
…named statuses (#1984) Per-model {model, status: ok|denied|unavailable|invalid|rate-limited|error, latency_ms, reason, credential_fp, checked_at} via Provider.getLanguage; credentials stay in the engine, reasons are redacted, credential_fp is a truncated domain-separated sha256. Classifier beside parseAPICallError. maxOutputTokens 1, 10s timeout, concurrency 4. Part of #1979.
Contributor
|
Important Draft PR not reviewedDraft PRs are not automatically reviewed by default.
To automatically review draft PRs, update your CodeRabbit configuration: reviews:
auto_review:
drafts: true
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
…ote routes, wired in extension and hub runner (#1985)
…esher probe wins, once-per-session gate, switch-target rules (#1989)
…valid turn failures, composer warning with one-click switch (#1989)
…by role, 12 cap, allowlist hard-filter incl. search, search reaches all shown; favorites persistence; model-picker config route (#1987)
…commended by role → recent, 12 cap, rest behind search; model.recommended/allowlist via GET /amicode/model-picker (#1987)
…ico/issue-1979-model-picker # Conflicts: # packages/app-bundle/manifest.json
…issing —, context K/M, tools, latency), cheapest-that-works search sort (dead never above ok, ties by name, no-query order untouched), dead-pick recommended from S4 config (#1988)
…, tools, measured latency inline + in tooltip; cheapest-that-works search sort (dead never above ok); dead-pick switch prefers S4 recommended (#1988)
…parsed, default 16, cap 64); wrapped Bedrock message retries too; both-rejected takes the second classification; floor error never invalid (#1994)
…from the message, default 16, cap 64) and classifies the retry; dedicated token-floor matcher ahead of invalid so the floor error is never `invalid` (#1994)
Member
Author
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
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.
Integration PR for #1979. Slices merged in DAG order:
POST /provider/probe. One-token probe returning named statuses; credentials stay in the engine.AMICODE_OPS_DIR, locked and shared by the shards. Scheduler andGET/refresh/demoteroutes, also wired into the hub runner.model.recommended/model.allowlistcome from$AMICODE_OPS_DIR/model-picker.jsonviaGET /amicode/model-picker.Director gates on the merged tree: typecheck, overlay typecheck and drift 0; bun 85/85 plus engine probe 9/9; service vitest 44/44.
Not merged until live-tested on the hub.