Commit Graph

5 Commits

Author SHA1 Message Date
daveadmin 8a11001bff Add AWS Bedrock three-tier gateway routing (LiteLLM via Colin)
Routes AI tools across three tiers based on task complexity:
- Azure GPT-4o-mini always: redact, translate, timeline-basic, search-legal (mechanical tasks)
- Claude Haiku 4.5 (Bedrock): ask, summarize, timeline-deep, citations (Norwegian nuance)
- Claude Sonnet 4.6 (Bedrock): korrespond, legal-analysis, deep-research, barnevernet-analyze,
  discrepancy-find, advocate (public-facing legal output)

No AWS credentials in app — credentials live in LiteLLM on Colin (same as nova-lite).
Rollback: DBN_BEDROCK_ENABLED=false in .env, no code push needed.

Includes extended thinking support for Pro deep-research via chatWithThinking().
Claude Opus 4.7 constant added for future premium tier (needs litellm_config.yaml entry).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-25 15:22:48 +02:00
daveadmin 17ad54cf36 Add chunked timeline routing 2026-05-25 12:34:41 +02:00
daveadmin 75b19f1dcf Add timeline engine size routing 2026-05-25 11:33:47 +02:00
daveadmin f00d3d68e5 Add Quick mode (nova-lite/Bedrock) as 3rd tier for timeline tool
Timeline now offers Quick/Standard/Deep: nova_lite routes to Amazon
Bedrock nova-lite via LiteLLM (1 credit, ~2s faster), azure_mini stays
gpt-4o-mini (1 credit), azure_full stays gpt-4o (2 credits, Pro only).
ToolModels tier rules: free→nova_lite only, plus→quick/standard,
pro→all three.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-25 10:26:07 +02:00
daveadmin 83fc71414f Add premium My Case MVP 2026-05-23 10:17:34 +02:00