bytelyst/learning_ai_common_plat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,350 Commits 2 Branches 3 Tags 60 MiB
c8b7861b2b
Commit Graph

4 Commits

Author SHA1 Message Date
root
39e48f3241 feat(llm): add FallbackLLMProvider + release pipeline script 
- packages/llm: add FallbackLLMProvider (providers/fallback.ts) that
  tries each provider in order, skipping unconfigured or erroring ones;
  wire 'fallback' as a first-class LLMProviderType in factory + types
- packages/llm: improve auto-detection in factory — PERPLEXITY_API_KEY
  and GEMINI_API_KEY trigger auto-selection when no explicit provider set
- scripts/release.sh: new pipeline — rebase from origin/main, build,
  apply changesets, publish outdated packages to Gitea registry, push
- scripts/run-registry-tests.sh: fix Gitea URL health-check to use a
  real package endpoint with auth header instead of bare registry root
- docs: mark Vercel track-B prompts B1–B3 as complete

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-13 04:21:12 +00:00
saravanakumardb1
151e07207b feat(llm): add vision, streaming, and embedding support 
- ContentPart types (TextContentPart, ImageUrlContentPart) for multipart messages
- ChatMessage.content now accepts string | ContentPart[] for vision
- EmbeddingRequest/Response types + optional embed() on LLMProvider
- chatCompletionStream() implemented in OpenAI + Azure providers (SSE parsing)
- embed() implemented in OpenAI + Azure providers
- Vision helpers: isVisionMessage, hasVisionContent, buildVisionMessage, getMessageText
- MockLLMProvider: streaming, embedding, vision content support
- 27 tests passing (up from 7)
 2026-04-06 07:42:30 -07:00
Saravana Achu Mac
43bf51a290 feat(llm): add Perplexity, Gemini providers and createFallbackChain 
- Add PerplexityProvider (OpenAI-compatible, reads PERPLEXITY_API_KEY)
- Add GeminiProvider (Google Generative Language API adapter, reads GEMINI_API_KEY)
- Add createFallbackChain() — ordered provider chain, skips unconfigured,
  aggregates errors; allows any app to replace custom LLM fallback loops
- Extend LLMProviderType with 'perplexity' | 'gemini'
- Update factory to resolve and instantiate new provider types
- Add PAID_PROVIDERS to llm-router registry (OpenAI, Perplexity) for apps
  using round-robin routing alongside free-tier providers
- 27 tests covering fallback chain, new providers, error/edge cases

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-04-05 12:50:08 -07:00
saravanakumardb1
dfa5eb73fa feat(packages): add cloud-agnostic abstraction packages — datastore, storage, llm, push + refactor secrets (58 tests) 2026-03-02 00:43:06 -08:00