ab044e6d6d
muse-spark-1.2-contributor-free returned HTTP 500 on /zen/v1/chat/completions.
The model is only served by /zen/v1/responses, so route it there via a per-model
targetFormat and normalize the Chat fields the Responses API rejects
(max_tokens -> max_output_tokens, reasoning_effort -> reasoning{effort,summary}),
clamping max/ultra down to the highest effort the model accepts (xhigh).
Routing stays per-model: the other free models (big-pickle, hy3-free, mimo,
nemotron, laguna) are not served by /responses and keep /chat/completions.
95 lines
3.1 KiB
JavaScript
95 lines
3.1 KiB
JavaScript
import { describe, expect, it } from "vitest";
|
|
import { getCapabilitiesForModel } from "../../open-sse/providers/capabilities.js";
|
|
import { PROVIDER_MODELS } from "../../open-sse/config/providerModels.js";
|
|
import { getThinkingLevels } from "../../open-sse/providers/thinkingLevels.js";
|
|
import { FORMATS } from "../../open-sse/translator/formats.js";
|
|
import { OpenCodeExecutor } from "../../open-sse/executors/opencode.js";
|
|
import "../translator/registerAll.js";
|
|
import { translateRequest } from "../../open-sse/translator/index.js";
|
|
|
|
const MODEL = "muse-spark-1.2-contributor-free";
|
|
const PROVIDER = "opencode";
|
|
|
|
const input = [{
|
|
type: "message",
|
|
role: "user",
|
|
content: [{ type: "input_text", text: "Think, then answer: 2 + 2?" }],
|
|
}];
|
|
|
|
describe("OpenCode Free Muse Spark thinking", () => {
|
|
it("advertises reasoning and the requested model limits", () => {
|
|
expect(PROVIDER_MODELS.oc?.some((model) => model.id === MODEL)).toBe(true);
|
|
expect(getCapabilitiesForModel(PROVIDER, MODEL)).toMatchObject({
|
|
reasoning: true,
|
|
thinkingFormat: "openai",
|
|
contextWindow: 1048576,
|
|
maxOutput: 131072,
|
|
});
|
|
expect(getCapabilitiesForModel(PROVIDER, `oc/${MODEL}`)).toMatchObject({
|
|
reasoning: true,
|
|
contextWindow: 1048576,
|
|
maxOutput: 131072,
|
|
});
|
|
expect(getThinkingLevels(PROVIDER, MODEL)).toEqual([
|
|
"none",
|
|
"minimal",
|
|
"low",
|
|
"medium",
|
|
"high",
|
|
"xhigh",
|
|
]);
|
|
});
|
|
|
|
it("clamps max to xhigh and emits the Responses reasoning shape", () => {
|
|
const body = {
|
|
input,
|
|
reasoning: { effort: "max" },
|
|
max_tokens: 131072,
|
|
};
|
|
|
|
const out = new OpenCodeExecutor().transformRequest(MODEL, body, true, {
|
|
connectionId: "opencode-muse-spark-test",
|
|
});
|
|
|
|
expect(out.reasoning).toEqual({ effort: "xhigh", summary: "auto" });
|
|
expect(out.reasoning_effort).toBeUndefined();
|
|
expect(out.max_output_tokens).toBe(131072);
|
|
expect(out.max_tokens).toBeUndefined();
|
|
});
|
|
|
|
it("leaves the other free models on Chat Completions", () => {
|
|
const executor = new OpenCodeExecutor();
|
|
const body = { messages: [{ role: "user", content: "hi" }], max_tokens: 1024 };
|
|
executor.transformRequest("big-pickle", body, true, {});
|
|
expect(executor.buildUrl("big-pickle")).toBe("https://opencode.ai/zen/v1/chat/completions");
|
|
expect(body.max_tokens).toBe(1024);
|
|
expect(body.max_output_tokens).toBeUndefined();
|
|
});
|
|
|
|
it("translates Chat Completions max thinking into a Responses request", () => {
|
|
const body = {
|
|
model: `oc/${MODEL}`,
|
|
messages: [{ role: "user", content: "Think, then answer: 2 + 2?" }],
|
|
reasoning_effort: "max",
|
|
max_tokens: 131072,
|
|
};
|
|
|
|
const translated = translateRequest(
|
|
FORMATS.OPENAI,
|
|
FORMATS.OPENAI_RESPONSES,
|
|
MODEL,
|
|
body,
|
|
true,
|
|
{},
|
|
PROVIDER,
|
|
);
|
|
const out = new OpenCodeExecutor().transformRequest(MODEL, translated, true, {
|
|
connectionId: "opencode-muse-spark-translation-test",
|
|
});
|
|
|
|
expect(out.reasoning).toEqual({ effort: "xhigh", summary: "auto" });
|
|
expect(out.max_output_tokens).toBe(131072);
|
|
expect(out.max_tokens).toBeUndefined();
|
|
});
|
|
});
|