Files
OmniRoute/open-sse/executors/petals.ts
Cong Vu Chi 15b4a54454 fix(claude): preserve tool_result adjacency (#1555)
* fix(claude): preserve tool_result adjacency in native and CC-compatible paths

* feat(providers): add Petals and Nous Research provider support

Register Nous Research as an OpenAI-compatible gateway with remote
model discovery and validation against chat completions.

Add Petals provider metadata, default config, validation, and a
specialized executor that maps OpenAI-style requests to the public
generate endpoint. Also allow optional API keys and configurable base
URLs for Petals in the dashboard and provider schemas.

Expand provider model and catalog tests to cover both integrations.

* fix(resilience): sync queue updates and clear stale discovery caches

Await runtime request queue updates so limiter settings and auto-enabled
API key protections are recomputed when resilience settings change.

Preserve cancelled batch state for in-flight work by marking input files
processed without generating output artifacts, and replace cached synced
models with an empty set when remote discovery returns no models so the
providers route falls back to the local catalog instead of stale cache.

---------

Co-authored-by: congvc <congvc-dev@gmail.com>
Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com>
2026-04-24 09:22:46 -03:00

386 lines
9.8 KiB
TypeScript

import { randomUUID } from "node:crypto";
import {
BaseExecutor,
mergeUpstreamExtraHeaders,
type ExecuteInput,
type ProviderCredentials,
} from "./base.ts";
import {
PETALS_DEFAULT_BASE_URL,
PETALS_DEFAULT_MODEL,
normalizePetalsBaseUrl,
} from "../config/petals.ts";
import { PROVIDERS } from "../config/constants.ts";
type JsonRecord = Record<string, unknown>;
type OpenAIMessage = {
role?: string;
content?: unknown;
};
function asRecord(value: unknown): JsonRecord {
return value && typeof value === "object" && !Array.isArray(value) ? (value as JsonRecord) : {};
}
function extractTextContent(content: unknown): string {
if (typeof content === "string") {
return content.trim();
}
if (!Array.isArray(content)) {
return "";
}
return content
.map((part) => {
if (!part || typeof part !== "object") return "";
const item = part as Record<string, unknown>;
if (item.type === "text" && typeof item.text === "string") {
return item.text;
}
if (item.type === "input_text" && typeof item.text === "string") {
return item.text;
}
return "";
})
.filter((text) => text.trim().length > 0)
.join("\n")
.trim();
}
function resolvePrompt(body: unknown): string {
const payload = asRecord(body);
const directPrompt = extractTextContent(payload.prompt);
if (directPrompt) {
return directPrompt;
}
const directInput = extractTextContent(payload.input);
if (directInput) {
return directInput;
}
const messages = Array.isArray(payload.messages) ? (payload.messages as OpenAIMessage[]) : [];
if (messages.length === 0) return "";
const systemParts: string[] = [];
const transcript: string[] = [];
let lastRole = "";
for (const message of messages) {
const role = String(message?.role || "user").toLowerCase();
const text = extractTextContent(message?.content);
if (!text) continue;
if (role === "system" || role === "developer") {
systemParts.push(text);
continue;
}
if (role === "assistant") {
transcript.push(`Assistant: ${text}`);
lastRole = "assistant";
continue;
}
transcript.push(`User: ${text}`);
lastRole = "user";
}
if (transcript.length === 0) {
return systemParts.join("\n\n").trim();
}
const parts: string[] = [];
if (systemParts.length > 0) {
parts.push(`System:\n${systemParts.join("\n\n")}`);
}
parts.push(transcript.join("\n\n"));
if (lastRole !== "assistant") {
parts.push("Assistant:");
}
return parts.join("\n\n").trim();
}
function resolveMaxNewTokens(body: unknown): number {
const payload = asRecord(body);
const candidates = [
payload.max_new_tokens,
payload.max_completion_tokens,
payload.max_output_tokens,
payload.max_tokens,
];
for (const value of candidates) {
if (typeof value === "number" && Number.isFinite(value) && value > 0) {
return Math.max(1, Math.min(4096, Math.floor(value)));
}
}
return 256;
}
function buildRequestPayload(model: string, body: unknown): URLSearchParams | null {
const payload = asRecord(body);
const prompt = resolvePrompt(payload);
if (!prompt) return null;
const form = new URLSearchParams();
form.set("model", model || PETALS_DEFAULT_MODEL);
form.set("inputs", prompt);
form.set("max_new_tokens", String(resolveMaxNewTokens(payload)));
const hasSampling =
typeof payload.temperature === "number" ||
typeof payload.top_k === "number" ||
typeof payload.top_p === "number";
if (hasSampling) {
form.set("do_sample", "1");
}
if (typeof payload.temperature === "number") {
form.set("temperature", String(payload.temperature));
}
if (typeof payload.top_k === "number") {
form.set("top_k", String(Math.max(1, Math.floor(payload.top_k))));
}
if (typeof payload.top_p === "number") {
form.set("top_p", String(payload.top_p));
}
if (typeof payload.repetition_penalty === "number") {
form.set("repetition_penalty", String(payload.repetition_penalty));
}
return form;
}
function estimateTokens(text: string): number {
return Math.max(1, Math.ceil(text.length / 4));
}
function buildSseChunk(data: unknown): string {
return `data: ${JSON.stringify(data)}\n\n`;
}
function buildOpenAiJsonCompletion(
content: string,
model: string,
id: string,
created: number
): Response {
const completionTokens = estimateTokens(content);
return new Response(
JSON.stringify({
id,
object: "chat.completion",
created,
model,
choices: [
{
index: 0,
message: { role: "assistant", content },
finish_reason: "stop",
},
],
usage: {
prompt_tokens: completionTokens,
completion_tokens: completionTokens,
total_tokens: completionTokens * 2,
},
}),
{
status: 200,
headers: { "Content-Type": "application/json" },
}
);
}
function buildSynthesizedStream(
content: string,
model: string,
id: string,
created: number
): Response {
const encoder = new TextEncoder();
const body = new ReadableStream<Uint8Array>({
start(controller) {
controller.enqueue(
encoder.encode(
buildSseChunk({
id,
object: "chat.completion.chunk",
created,
model,
choices: [{ index: 0, delta: { role: "assistant" }, finish_reason: null }],
})
)
);
if (content) {
controller.enqueue(
encoder.encode(
buildSseChunk({
id,
object: "chat.completion.chunk",
created,
model,
choices: [{ index: 0, delta: { content }, finish_reason: null }],
})
)
);
}
controller.enqueue(
encoder.encode(
buildSseChunk({
id,
object: "chat.completion.chunk",
created,
model,
choices: [{ index: 0, delta: {}, finish_reason: "stop" }],
})
)
);
controller.enqueue(encoder.encode("data: [DONE]\n\n"));
controller.close();
},
});
return new Response(body, {
status: 200,
headers: { "Content-Type": "text/event-stream" },
});
}
function toOpenAiError(status: number, message: string): Response {
return new Response(
JSON.stringify({
error: {
message,
type:
status === 401 || status === 403
? "authentication_error"
: status === 429
? "rate_limit_error"
: "api_error",
},
}),
{
status,
headers: { "Content-Type": "application/json" },
}
);
}
export class PetalsExecutor extends BaseExecutor {
constructor() {
super("petals", PROVIDERS.petals || { format: "openai", baseUrl: PETALS_DEFAULT_BASE_URL });
}
buildUrl(
_model: string,
_stream: boolean,
_urlIndex = 0,
credentials: ProviderCredentials | null = null
): string {
const rawBaseUrl =
typeof credentials?.providerSpecificData?.baseUrl === "string"
? credentials.providerSpecificData.baseUrl
: this.config.baseUrl;
return normalizePetalsBaseUrl(rawBaseUrl);
}
buildHeaders(credentials: ProviderCredentials | null): Record<string, string> {
const token = credentials?.apiKey || credentials?.accessToken;
return {
"Content-Type": "application/x-www-form-urlencoded",
Accept: "application/json",
...(token ? { Authorization: `Bearer ${token}` } : {}),
};
}
async execute({ model, body, stream, credentials, signal, upstreamExtraHeaders }: ExecuteInput) {
const resolvedModel = model || PETALS_DEFAULT_MODEL;
const payload = buildRequestPayload(resolvedModel, body);
const url = this.buildUrl(resolvedModel, stream, 0, credentials);
const headers = this.buildHeaders(credentials);
mergeUpstreamExtraHeaders(headers, upstreamExtraHeaders);
if (!payload) {
return {
response: toOpenAiError(400, "Petals requests require at least one user prompt."),
url,
headers,
transformedBody: body,
};
}
const transformedBody = Object.fromEntries(payload.entries());
try {
const response = await fetch(url, {
method: "POST",
headers,
body: payload.toString(),
signal,
});
if (!response.ok) {
const errorText = await response.text();
return {
response: toOpenAiError(
response.status,
`Petals API failed with status ${response.status}: ${errorText || "Unknown error"}`
),
url,
headers,
transformedBody,
};
}
const json = asRecord(await response.json());
if (json.ok === false) {
const traceback =
typeof json.traceback === "string" && json.traceback.trim()
? json.traceback.trim()
: "Unknown Petals upstream error";
return {
response: toOpenAiError(502, `Petals API error: ${traceback}`),
url,
headers,
transformedBody,
};
}
const content = typeof json.outputs === "string" ? json.outputs : "";
const id = `chatcmpl-petals-${randomUUID()}`;
const created = Math.floor(Date.now() / 1000);
return {
response: stream
? buildSynthesizedStream(content, resolvedModel, id, created)
: buildOpenAiJsonCompletion(content, resolvedModel, id, created),
url,
headers,
transformedBody,
};
} catch (error) {
const message = error instanceof Error ? error.message : String(error || "Unknown error");
return {
response: toOpenAiError(502, `Petals fetch error: ${message}`),
url,
headers,
transformedBody,
};
}
}
}
export default PetalsExecutor;