feat: add chat cost analytics backend (#23036)

Add cost tracking for LLM chat interactions with microdollar precision.

## Changes
- Add `chatcost` package for per-message cost calculation using
`shopspring/decimal` for intermediate arithmetic
- **Ceil rounding policy**: fractional micros round UP to next whole
micro (applied once after summing all components)
- Database migration: `total_cost_micros` BIGINT column with historical
backfill and `created_at` index
- API endpoints: per-user cost summary and admin rollup under
`/api/experimental/chats/cost/`
- SDK types: `ChatCostSummary`, `ChatCostModelBreakdown`,
`ChatCostUserRollup`
- Fix `modeloptionsgen` to handle `decimal.Decimal` as opaque numeric
type
- Update frontend pricing test fixtures for string decimal types

## Design decisions
- `NULL` = unpriced (no matching model config), `0` = free
- Reasoning tokens included in output tokens (no double-counting)
- Integer microdollars (BIGINT) for storage and API responses
- Price config uses `decimal.Decimal` for exact parsing; totals use
`int64`

Frontend: #23037
This commit is contained in:
Michael Suchacz
2026-03-13 18:30:49 +01:00
committed by GitHub
parent 1152b61ebb
commit c3b6284955
34 changed files with 2034 additions and 262 deletions
+13
View File
@@ -215,6 +215,19 @@ type sqlcQuerier interface {
GetAuthorizationUserRoles(ctx context.Context, userID uuid.UUID) (GetAuthorizationUserRolesRow, error)
GetChatByID(ctx context.Context, id uuid.UUID) (Chat, error)
GetChatByIDForUpdate(ctx context.Context, id uuid.UUID) (Chat, error)
// Per-root-chat cost breakdown for a single user within a date range.
// Groups by root_chat_id so forked chats roll up under their root.
// Only counts assistant-role messages.
GetChatCostPerChat(ctx context.Context, arg GetChatCostPerChatParams) ([]GetChatCostPerChatRow, error)
// Per-model cost breakdown for a single user within a date range.
// Only counts assistant-role messages that have a model_config_id.
GetChatCostPerModel(ctx context.Context, arg GetChatCostPerModelParams) ([]GetChatCostPerModelRow, error)
// Deployment-wide per-user cost rollup within a date range.
// Only counts assistant-role messages.
GetChatCostPerUser(ctx context.Context, arg GetChatCostPerUserParams) ([]GetChatCostPerUserRow, error)
// Aggregate cost summary for a single user within a date range.
// Only counts assistant-role messages.
GetChatCostSummary(ctx context.Context, arg GetChatCostSummaryParams) (GetChatCostSummaryRow, error)
GetChatDiffStatusByChatID(ctx context.Context, chatID uuid.UUID) (ChatDiffStatus, error)
GetChatDiffStatusesByChatIDs(ctx context.Context, chatIds []uuid.UUID) ([]ChatDiffStatus, error)
GetChatFileByID(ctx context.Context, id uuid.UUID) (ChatFile, error)