Skip to content

chore(ai-gateway): prune delisted unavailable models #5783

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
chrarnoldus wants to merge 1 commit into
base: main
Choose a base branch
from
Open

chore(ai-gateway): prune delisted unavailable models #5783

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions apps/web/src/app/api/openrouter/models/[provider]/[model]/endpoints/route.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,10 +89,10 @@ describe('GET /api/openrouter/models/[provider]/[model]/endpoints', () => {

test('returns 404 for unavailable models without reading cached metadata', async () => {
mockedGetOpenRouterModelsMetadataFromDatabase.mockClear();
const modelId = 'openai/gpt-oss-20b:free';
const modelId = 'google/gemma-4-31b-it:free';

const response = await GET(request(modelId), {
params: Promise.resolve({ provider: 'openai', model: 'gpt-oss-20b:free' }),
params: Promise.resolve({ provider: 'google', model: 'gemma-4-31b-it:free' }),
});

expect(response.status).toBe(404);
Expand Down
6 changes: 3 additions & 3 deletions apps/web/src/lib/ai-gateway/providers/openrouter/free-endpoint-data-policy.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -217,14 +217,14 @@ describe('applyFreeEndpointDataPolicy', () => {
});

test('ignores unavailable OpenRouter and exclusive free models', () => {
const freeVariant = offering('openai/gpt-oss-20b', true);
const model = offering('openai/gpt-oss-20b');
const freeVariant = offering('google/gemma-4-31b-it', true);
const model = offering('google/gemma-4-31b-it');
const providerModelData = [{ provider: { slug: 'darkbloom' }, models: [freeVariant, model] }];

applyFreeEndpointDataPolicy({
providerModelData,
openRouterFreeEndpoints: getOpenRouterFreeEndpoints(providerModelData),
kiloExclusiveModels: [freeExclusiveModel('openai/gpt-oss-20b:free', [])],
kiloExclusiveModels: [freeExclusiveModel('google/gemma-4-31b-it:free', [])],
});

expect(freeVariant.endpoint?.data_policy).toEqual({ training: false, retainsPrompts: false });
Expand Down
27 changes: 19 additions & 8 deletions apps/web/src/lib/ai-gateway/unavailable-models.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,19 +6,30 @@ import {

describe('unavailable models', () => {
test('keeps exact matching for request rejection', () => {
expect(isUnavailableModel('openai/gpt-oss-20b:free')).toBe(true);
expect(isUnavailableModel('google/gemma-4-26b-a4b-it:free')).toBe(true);
expect(isUnavailableModel('sakana/fugu-ultra')).toBe(false);
expect(isUnavailableModel('google/gemma-4-31b-it:free')).toBe(true);
expect(isUnavailableModel('openai/gpt-oss-20b')).toBe(false);
expect(isUnavailableModel('tencent/hy3:free')).toBe(true);
expect(isUnavailableModel('tencent/hy3')).toBe(false);
expect(isUnavailableModel('google/gemma-4-26b-a4b-it')).toBe(false);
expect(isUnavailableModel('z-ai/glm-5.2:free')).toBe(true);
expect(isUnavailableModel('z-ai/glm-5.2')).toBe(false);
});

test('matches normalized families for provider metadata', () => {
expect(familyHasUnavailableFreeModel('openai/gpt-oss-20b:free')).toBe(true);
expect(familyHasUnavailableFreeModel('openai/gpt-oss-20b')).toBe(true);
expect(familyHasUnavailableFreeModel('google/gemma-4-26b-a4b-it:free')).toBe(true);
expect(familyHasUnavailableFreeModel('google/gemma-4-26b-a4b-it')).toBe(true);
expect(familyHasUnavailableFreeModel('cohere/north-mini-code')).toBe(false);
expect(familyHasUnavailableFreeModel('tencent/hy3:free')).toBe(true);
expect(familyHasUnavailableFreeModel('tencent/hy3')).toBe(true);
expect(familyHasUnavailableFreeModel('z-ai/glm-5.2:free')).toBe(true);
expect(familyHasUnavailableFreeModel('z-ai/glm-5.2')).toBe(true);
});

test.each([
'nvidia/nemotron-3-nano-30b-a3b:free',
'nvidia/nemotron-nano-12b-v2-vl:free',
'nvidia/nemotron-nano-9b-v2:free',
'openai/gpt-oss-20b:free',
'tencent/hy3:free',
])('does not explicitly block delisted model %s', modelId => {
expect(isUnavailableModel(modelId)).toBe(false);
expect(familyHasUnavailableFreeModel(modelId)).toBe(false);
});
});
5 changes: 0 additions & 5 deletions apps/web/src/lib/ai-gateway/unavailable-models.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,11 +3,6 @@ import { normalizeModelId } from '@/lib/ai-gateway/model-utils';
const unavailableModelIds: ReadonlySet<string> = new Set([
'google/gemma-4-26b-a4b-it:free', // usable through kilo-auto
'google/gemma-4-31b-it:free',
'nvidia/nemotron-3-nano-30b-a3b:free',
'nvidia/nemotron-nano-12b-v2-vl:free',
'nvidia/nemotron-nano-9b-v2:free',
'openai/gpt-oss-20b:free',
'tencent/hy3:free',
'z-ai/glm-5.2:free',
]);

Expand Down