pi-mono/packages/ai/test/anthropic-adaptive-thinking-models.test.ts
David Brailovsky d2e7d8b7c1 feat(ai): add MiniMax-M3.1-Flash and hide only old MiniMax models
Replace the direct minimax allowlist with a denylist of the older M2, M2.1,
M2.5, and M2.5-highspeed models, applied where MiniMax models are built.
MiniMax models with models.dev effort levels use adaptive thinking with
output_config.effort and mark thinking off as unsupported, since
MiniMax-M3.1-Flash rejects thinking.type=disabled.

closes #11121
2026-10-19 11:35:34 +02:00

50 lines
2 KiB
TypeScript

import { describe, expect, it } from "vitest";
import { getModels, getProviders } from "../src/compat.ts";
import type { Api, Model } from "../src/types.ts";
const EXPECTED_CURRENT_ADAPTIVE_THINKING_MODELS = [
"anthropic/claude-fable-5",
"anthropic/claude-haiku-5-5",
"anthropic/claude-opus-4-8",
"anthropic/claude-opus-5",
"anthropic/claude-sonnet-5",
"cloudflare-ai-gateway/claude-fable-5",
"fireworks/accounts/fireworks/models/deepseek-v4p1-flash",
"fireworks/accounts/fireworks/models/gpt-oss-120b",
"fireworks/accounts/fireworks/models/qwen3p8-max",
"minimax-cn/MiniMax-M3.1-Flash",
"minimax/MiniMax-M3.1-Flash",
"opencode/claude-opus-4-8",
"opencode/claude-opus-5",
"vercel-ai-gateway/anthropic/claude-opus-4.8",
"vercel-ai-gateway/anthropic/claude-opus-5",
"vercel-ai-gateway/anthropic/claude-sonnet-5",
];
function getAllModels(): Model<Api>[] {
return getProviders().flatMap((provider) => getModels(provider) as Model<Api>[]);
}
describe("Anthropic adaptive thinking model metadata", () => {
it("marks built-in Anthropic Messages models that use adaptive thinking", () => {
const flaggedModels = getAllModels()
.filter((model): model is Model<"anthropic-messages"> => model.api === "anthropic-messages")
.filter((model) => model.compat?.forceAdaptiveThinking === true)
.map((model) => `${model.provider}/${model.id}`)
.sort();
expect(flaggedModels).toEqual(expect.arrayContaining([...EXPECTED_CURRENT_ADAPTIVE_THINKING_MODELS].sort()));
expect(flaggedModels).toEqual(
flaggedModels.filter(
(modelId) =>
// Regression for #9323: Fireworks uses catalog effort metadata and
// verified fallbacks, not a fixed set of adaptive model names.
modelId.startsWith("fireworks/") ||
// #11121: MiniMax models with catalog effort levels require adaptive thinking.
modelId.startsWith("minimax/") ||
modelId.startsWith("minimax-cn/") ||
/(opus[-.](4[-.][678]|5)|sonnet[-.]4[-.]6|sonnet[-.]5|haiku[-.]5[-.]5|fable[-.]5)/.test(modelId),
),
);
});
});