>_
.issue.db
/highway
Dashboard
Issues
Memory
Lessons
Audit Log
New Issue
Edit Issue #1065
Update issue details
Title *
Description
EARS SPEC: - When an LLM call is served by provider prism, the engine shall record a non-zero estimated USD cost from Prism's published per-token price for that model. - If a paid provider/model has no price entry, then the engine shall make the gap visible in the health or usage API, not only in a once-per-process log line. FOUND 2026-10-25 01:00Z after the 4b3079f rollout (#1060): worker log 'No pricing entry for prism/deepseek-v4-flash - recording cost as $0.0'. engine/tools/llm/usage.py PRICING has no 'prism' key. Every chat route is prism since #1044, so all agent usage now reports $0. [llm.budget] max_usd_per_run = 0 in docker/config.ini, so the USD rail is off by configuration today; max_tokens_per_run still applies. Needs Prism's per-token price for deepseek-v4-flash (ask prism-dev).
Priority
Low
Medium
High
Critical
Status
Open
In Progress
Closed
Won't Do
Due Date (YYYY-MM-DD)
Tags (comma separated)
Related Issues (IDs)
Enter IDs of issues related to this one. They will be linked as 'related'.
Update Issue
Cancel