Price a request from plain values instead of the plugin request context
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
调研方向
Read core/cost/settle/settle.go and core/cost/settle/avoided.go, then inspect core/plugins/inferenceparser/cost.go for the current callers. Add table tests for plain Input pricing, LiteLLM cost headers, and Comparable discount and margin checks. Done means Settle and Comparable accept plain values, their context forms wrap them, and existing cost tests pass unchanged.
由索引模型根据 Issue 内容生成。
描述
Part of #1437 (assessment: laptop data plane on Praxis) · prep work, useful to Cortex on its own
Status: there is no shipped Cortex–Praxis integration today (the paused
cmd/cortex-praxisconverter is not one), and the interim path is not decided (LLM traffic to praxis-ai as an inference server, or all traffic through Praxis via #1491). Nothing here depends on that choice: each item is useful to Cortex now, and would make a possible move to Praxis easier whichever path is taken.
Problem
Pricing a request needs the whole plugin request context.
settle.Settle(pctx, rates)readspctx.Host,pctx.Extensions.Inferenceandpctx.ResponseHeaders(LiteLLM cost headers, andContent-TypeviaIsEventStream);pctx.Pathonly in log lines.Avoidedreadspctx.Extensions.Custom(tool-prune's record), the request body andBodyMutated().Publishwrites the result back intopctxunder two keys;Storekeeps it in pipeline state, which litellm-budget-track reads withLoadand rewrites withAmend.
Eight exported functions take *pipeline.Context: Settle, Avoided, IsEventStream, Comparable, Store, Load, Publish, Amend.
Settle is called only from inside the inference-parser plugin (cost.go:72-112), which also calls Store, Avoided and IsEventStream; litellm-budget-track calls Comparable, Load and Amend. So pricing can run only inside Cortex's Go pipeline: not from a recorded event, not in a test without building a pctx, not from another proxy.
Proposal
This issue owns settle's input. #1497 item 3 owns recording the cost-relevant headers on the event, and #1503 depends on this one.
Settle(in Input, rates) SettledwithInput{Host, Inference, ResponseHeaders}.IsEventStreamis derived fromResponseHeaders;Pathis only for logging and can be passed separately or dropped.Comparable(in Input)the same way.Avoidedkeeps apctxform for now: it needs tool-prune's record, the forwarded body andBodyMutated(), none of which belong in a pricing input.Store,Load,PublishandAmendare how plugins exchange the result; they stay onpctx(their serialization is #1499 item 4).- Keep
Settle(pctx, rates)as a thin wrapper over the new form, so existing callers and tests don't change. No behaviour change. - Where the wrapper lives (the inference-parser plugin, or a separate cost step) is decided in #1503.
Done when
SettleandComparablehave plain-value forms, and thepctxversions are wrappers over them.- A table test prices from plain
Inputvalues, including the LiteLLM cost-header cases and theComparablechecks on the discount and margin headers. (Gateway multipliers are on the rates side,pricing/multiplier.go, and are unaffected.) - Existing cost tests pass unchanged (about 43 test calls use
Settle(pctx, …)).
Evidence read at main @ e455a04 (2026-10-25); not built.
- 主要语言
- Go
- 星标
- 21
- 派生
- 43
- 平均合并
- 13 小时 8 分钟
- 30 天内合并 PR
- 267
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
rossoctl/cortex 的其他 Issue
-
难度 2/5 1-3 小时 新手友好度 72/100
维护者通常 1 天内回复
-
UI shows costs for all sessions even when drilled into a session可能已有人在做 @esnible 于 16 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 70/100
rossoctl/cortex#1116 · 2 条评论 · 已指派 1 人 ·
维护者通常 1 天内回复
-
难度 3/5 半天 新手友好度 70/100
维护者通常 1 天内回复
-
难度 4/5 一周以上 新手友好度 18/100
维护者通常 1 天内回复
-
难度 3/5 1-2 天 新手友好度 45/100
维护者通常 1 天内回复
相似的 Issue
-
ux
难度 2/5 1-3 小时 新手友好度 68/100
维护者通常 1 天内回复
-
bug report
难度 2/5 1-3 小时 新手友好度 70/100
维护者通常 1 天内回复
-
type:bug
难度 2/5 1-3 小时 新手友好度 66/100
nightgauge/nightgauge#2732 ·
维护者通常 1 天内回复
-
fix below CVEs未关闭bug
难度 2/5 1-3 小时 新手友好度 72/100
维护者通常 5 天内回复
-
good first issue help wanted
难度 2/5 1-3 小时 新手友好度 78/100