tben
d9aea8e9c8
Nothing anywhere recorded what a model call cost: the harness used a provider's usage numbers only to decide whether a turn had finished, then dropped them. So a reply could not say what it took to make, one model could not be compared with another on price, and there was no way to ask what a day, a week or a month of the crew's work came to. - `token_usages` records one row per call: user, bot, workspace, session, job and message, the harness, the provider and model that actually answered, the prompt and completion counts, and the cost when the provider reports one. - `harness-usage-scope.ts` carries those ids in an AsyncLocalStorage scope that `run-bot-job.ts` enters once, so a run's turns, its session summary, its transcriptions, images and embeddings are all attributed without threading ids through every harness function. Nested runs merge rather than replace, and a null genuinely clears a field (a bot's own chat has no workspace). - Work with no run behind it is charged to its owner instead: `run-with-fallback-scope.ts` reaches the workspace's creator else the bot's, and deliberately does not override a real run, so an ingest a bot asked for is not misattributed to the bot's owner. - All six harnesses record through `withCandidates`'s `on_success`. Recording never throws: a model that answered must not lose its answer to a failed insert. - A job's calls land before its message exists, so `complete-bot-job.ts` back-fills `message_id` once the message is inserted. - The message's actions row shows what it cost, beside the timestamp and the play button, with the breakdown in its title. - `/admin/usage` totals the install's spending for 7, 30 or 90 days: the total against the previous period, a day-by-day chart, a split per kind of call, and a table per model with the price per thousand tokens worked out from cost and tokens, which is what makes two models comparable. Crew managers only, since it is the whole install's work rather than one user's conversation. It sits with Routines and Activity in the user popover rather than under Settings, because it is a record of what happened, not something to configure. Days are grouped in local time (`date(created_at / 1000, 'unixepoch', 'localtime')`), matching the local midnights the window builds and the `YYYY-MM-DD` buckets it looks up. Grouping by the bare UTC day put a late evening's calls under a name no bucket carried, so they vanished and the chart's total silently disagreed with the totals card; a test pins it. `canSeeRailItem` now decides who sees a popover or rail entry. The two each filtered with their own inline rule and only one understood `can_view`, so a gated page could be offered in the popover to everyone.
100 lines
3.1 KiB
TypeScript
100 lines
3.1 KiB
TypeScript
import grabHarnessSetting from "@/src/functions/backend/db/harness/grab-harness-setting";
|
|
import grabHarnessAdapter from "../providers/grab-harness-adapter";
|
|
import resolveHarnessCandidates from "../resolve/resolve-harness-candidates";
|
|
import withCandidates from "../resolve/with-candidates";
|
|
import type { HarnessAudioStream, HarnessCandidate } from "../types";
|
|
import buildUsageCall from "../usage/build-usage-call";
|
|
import recordHarnessUsage from "../usage/record-harness-usage";
|
|
|
|
type Params = {
|
|
text: string;
|
|
/**
|
|
* `mp3` unless set.
|
|
*/
|
|
format?: string;
|
|
signal?: AbortSignal;
|
|
/**
|
|
* Uses this bot's speech model and voice when it has its own.
|
|
*/
|
|
bot_id?: number | string;
|
|
/**
|
|
* The workspace the call is in: its models win over the bot's.
|
|
*/
|
|
workspace_id?: number | string | null;
|
|
};
|
|
|
|
/**
|
|
* Reads text aloud with the speech harness, as a stream that can play
|
|
* while the rest is still being made. Falls back to the next model only
|
|
* until one answers. Uses the voice picked in the speech setting when the
|
|
* model has it, else the model's first voice.
|
|
*/
|
|
export default async function speakText({
|
|
text,
|
|
format,
|
|
signal,
|
|
bot_id,
|
|
workspace_id,
|
|
}: Params) {
|
|
const [candidates, setting] = await Promise.all([
|
|
resolveHarnessCandidates({ harness: "speech", bot_id, workspace_id }),
|
|
grabHarnessSetting({ harness: "speech", bot_id, workspace_id }),
|
|
]);
|
|
|
|
const { result } = await withCandidates<HarnessAudioStream>({
|
|
harness: "speech",
|
|
candidates,
|
|
/**
|
|
* Speech is billed per character and streamed, so there is no
|
|
* completion count to read: the call is recorded as it is, with
|
|
* the text's length standing in for what was spoken.
|
|
*/
|
|
on_success: async ({ candidate }) =>
|
|
await recordHarnessUsage({
|
|
call: buildUsageCall({
|
|
harness: "speech",
|
|
candidate,
|
|
usage: { prompt_tokens: text.length, completion_tokens: 0 },
|
|
}),
|
|
}),
|
|
run: async (candidate) =>
|
|
await grabHarnessAdapter({ kind: candidate.provider.kind }).speak({
|
|
candidate,
|
|
text,
|
|
voice: pickVoice({
|
|
candidate,
|
|
voice:
|
|
typeof setting.options.voice == "string"
|
|
? setting.options.voice
|
|
: undefined,
|
|
}),
|
|
format,
|
|
signal: signal || new AbortController().signal,
|
|
}),
|
|
});
|
|
|
|
return result;
|
|
}
|
|
|
|
/**
|
|
* The voice to ask for: the picked one when the model has it (or the
|
|
* model was picked by hand), else the model's first, else OpenAI's.
|
|
*/
|
|
export function pickVoice({
|
|
candidate,
|
|
voice,
|
|
}: {
|
|
candidate: HarnessCandidate;
|
|
voice?: string;
|
|
}) {
|
|
const voices = candidate.model.voices;
|
|
|
|
if (voice && (candidate.source == "pinned" || voices?.includes(voice))) {
|
|
return voice;
|
|
}
|
|
|
|
if (voices?.[0]) return voices[0];
|
|
|
|
return candidate.provider.kind == "openai" ? "alloy" : undefined;
|
|
}
|