Adds the Desktop-exclusive Preview models and the Xiaomi account-session
route to the existing xiaomi-mimo provider instead of a separate
xiaomi-desktop provider, so the dashboard shows one MiMo entry rather than
three overlapping ones.
Dual auth, same pattern as kimi — API key (sk-) covers the cloud API,
Desktop/OAuth adds the account session used by the Preview models:
- registry: category oauth, authModes [oauth, apikey], oauth block, the two
mimo-x-*-preview models, and the invite signupUrl
- executor: routes Preview models to the account-service route with a Cookie
session, everything else keeps the sourceFormat-matched transport
- oauth: custom ECDH encrypted-callback flow (X25519 -> SHA256 -> AES-256-GCM)
with a loopback callback proxy, plus one-click import of the local Desktop
auth.json
- usage: weekly quota from the account session
Fixes found while merging:
- the OAuth browser flow was dead: poll-status cleared the session before the
client could POST /exchange, so every exchange returned 400
- a Claude-format client was sent to /v1/chat/completions instead of the
declared /anthropic/v1/messages transport, because buildUrl ignored
runtimeTransport
- stopXiaomiMimoProxy leaked every pending session (each holding an X25519
private key) for the process lifetime
- the OAuth exchange did not persist the Desktop passToken, so the Preview
models could never work after a browser sign-in
Removes dead code: the local engine token minting (mimoEngine, never called
on the request path), the model-catalog and usage routes, engineToken/
engineUrl plumbing, and an unread top-level usage block.
Adds tests/unit/xiaomi-mimo-{executor,oauth-session,oauth-proxy}.test.js —
the provider previously had none.
79 lines
3.8 KiB
JavaScript
79 lines
3.8 KiB
JavaScript
import { getCapabilitiesForModel } from "../../providers/capabilities.js";
|
|
|
|
// Strip request params a given provider/model rejects upstream (e.g. HTTP 400).
|
|
// Config-driven: add a rule instead of scattering `delete body.x` across executors.
|
|
|
|
// Each rule: optional provider, regex match on model, list of params to drop.
|
|
// A param is removed only when it is present (!== undefined).
|
|
const STRIP_RULES = [
|
|
// All Claude models: temperature deprecated/rejected upstream (Anthropic 400). #1748
|
|
{ match: /claude/i, drop: ["temperature"] },
|
|
// GitHub Copilot gpt-5.4: temperature unsupported.
|
|
{ provider: "github", match: /gpt-5\.4/i, drop: ["temperature"] },
|
|
// GitHub Copilot Claude (except opus/sonnet 4.6): thinking + reasoning_effort rejected. #713
|
|
{ provider: "github", match: (m) => /claude/i.test(m) && !/claude.*(opus|sonnet).*4\.6/i.test(m), drop: ["thinking", "reasoning_effort"] },
|
|
// Cloudflare Workers AI: content must be plain string, rejects OpenAI content-part array (#1926)
|
|
{ provider: "cloudflare-ai", flattenContent: true },
|
|
// MiMo Desktop Preview models (account-service route): content must be plain string,
|
|
// rejects OpenAI content-part array. Cloud models keep their parts (mimo-v2-omni is multi-modal).
|
|
{ provider: "xiaomi-mimo", match: /preview/i, flattenContent: true },
|
|
{ provider: "volcengine-ark", match: /glm-5/i, clampToModelMaxOutput: true },
|
|
// VolcEngine Ark caps the Kimi family at max_tokens <= 32768, but the model's
|
|
// advertised ceiling is far higher (Kimi-K2.7-Code resolves to maxOutput 262144),
|
|
// so clampToModelMaxOutput alone leaves it uncapped and the request 400s with
|
|
// "integer above maximum value, expected <= 32768". Pin an explicit endpoint cap;
|
|
// min() with the model ceiling still applies if a variant's own limit is lower.
|
|
{ provider: "volcengine-ark", match: /kimi/i, maxOutputCap: 32768, clampToModelMaxOutput: true },
|
|
];
|
|
|
|
// Test a rule's match (regex or predicate) against the model id.
|
|
function matches(rule, model) {
|
|
if (!rule.match) return true;
|
|
return typeof rule.match === "function" ? rule.match(model) : rule.match.test(model);
|
|
}
|
|
|
|
function clampNumber(body, key, ceiling) {
|
|
if (typeof body[key] === "number" && Number.isFinite(body[key]) && body[key] > ceiling) {
|
|
body[key] = ceiling;
|
|
}
|
|
}
|
|
|
|
// Remove unsupported params from body in place; returns body.
|
|
export function stripUnsupportedParams(provider, model, body) {
|
|
if (!model || !body || typeof body !== "object") return body;
|
|
for (const rule of STRIP_RULES) {
|
|
if (rule.provider && rule.provider !== provider) continue;
|
|
if (!matches(rule, model)) continue;
|
|
for (const key of rule.drop || []) {
|
|
if (body[key] !== undefined) delete body[key];
|
|
}
|
|
// CF Workers AI oneOf root schema only accepts content as plain string (#1926)
|
|
if (rule.flattenContent && Array.isArray(body.messages)) {
|
|
for (const msg of body.messages) {
|
|
if (msg && Array.isArray(msg.content)) {
|
|
msg.content = msg.content
|
|
.map(b => (b?.type === "text" && typeof b.text === "string") ? b.text : "")
|
|
.join("");
|
|
}
|
|
}
|
|
}
|
|
if (rule.clampToModelMaxOutput || Number.isFinite(rule.maxOutputCap)) {
|
|
const modelCeiling = getCapabilitiesForModel(provider, model).maxOutput;
|
|
const candidates = [];
|
|
if (rule.clampToModelMaxOutput && Number.isFinite(modelCeiling) && modelCeiling > 0) {
|
|
candidates.push(modelCeiling);
|
|
}
|
|
if (Number.isFinite(rule.maxOutputCap) && rule.maxOutputCap > 0) {
|
|
candidates.push(rule.maxOutputCap);
|
|
}
|
|
if (candidates.length > 0) {
|
|
const ceiling = Math.min(...candidates);
|
|
clampNumber(body, "max_tokens", ceiling);
|
|
clampNumber(body, "max_completion_tokens", ceiling);
|
|
clampNumber(body, "max_output_tokens", ceiling);
|
|
}
|
|
}
|
|
}
|
|
return body;
|
|
}
|