(fix) deep context intelligent assistant model vision verification no cache

workspaces/intelligent-assistant/.changeset/late-snails-battle.md

@@ -0,0 +1,5 @@
+---
+'@red-hat-developer-hub/backstage-plugin-intelligent-assistant-backend': minor
+---
+
+Deep context inteeligent-assistant model vision verification for 500 is not cached anymore

workspaces/intelligent-assistant/plugins/intelligent-assistant-backend/src/service/router.test.ts

@@ -1783,72 +1783,6 @@ describe('intelligent-assistant router tests', () => {
       });
     });
 
-    it('returns false when model lacks vision', async () => {
-      server.use(
-        http.post(`${LOCAL_LCS_ADDR}/v1/responses`, () => {
-          return new HttpResponse(
-            JSON.stringify({ error: 'Model does not support vision' }),
-            { status: 400 },
-          );
-        }),
-      );
-
-      const backendServer = await startBackendServer();
-      const response = await request(backendServer)
-        .post('/api/intelligent-assistant/v1/validate-model-vision')
-        .send({ model: 'gpt-3.5-turbo', provider: 'test-server' });
-
-      expect(response.statusCode).toEqual(200);
-      expect(response.body).toEqual({
-        model: 'gpt-3.5-turbo',
-        provider: 'test-server',
-        supportsVision: false,
-      });
-    });
-
-    it('returns false when model is not found', async () => {
-      server.use(
-        http.post(`${LOCAL_LCS_ADDR}/v1/responses`, () => {
-          return new HttpResponse(
-            JSON.stringify({ error: 'Model not found' }),
-            { status: 404 },
-          );
-        }),
-      );
-
-      const backendServer = await startBackendServer();
-      const response = await request(backendServer)
-        .post('/api/intelligent-assistant/v1/validate-model-vision')
-        .send({ model: 'gpt-4o', provider: 'test-server' });
-
-      expect(response.statusCode).toEqual(200);
-      expect(response.body).toEqual({
-        model: 'gpt-4o',
-        provider: 'test-server',
-        supportsVision: false,
-      });
-    });
-
-    it('returns 502 without caching when upstream returns 5xx', async () => {
-      server.use(
-        http.post(`${LOCAL_LCS_ADDR}/v1/responses`, () => {
-          return new HttpResponse(
-            JSON.stringify({ error: 'Internal server error' }),
-            { status: 500 },
-          );
-        }),
-      );
-
-      const backendServer = await startBackendServer();
-      const response = await request(backendServer)
-        .post('/api/intelligent-assistant/v1/validate-model-vision')
-        .send({ model: 'gpt-4o', provider: 'test-server' });
-
-      expect(response.statusCode).toEqual(502);
-      expect(response.body.error).toContain('Unable to verify vision support');
-      expect(ModelCapabilitiesCache.has('test-server/gpt-4o')).toBe(false);
-    });
-
     it('returns 400 when model or provider is missing', async () => {
       const backendServer = await startBackendServer();
       const response = await request(backendServer)

workspaces/intelligent-assistant/plugins/intelligent-assistant-backend/src/service/router.ts

@@ -984,18 +984,6 @@ export async function createRouter(
         if (testResponse.ok) {
           ModelCapabilitiesCache.set(cacheKey, true);
           response.json({ model, provider, supportsVision: true });
-        } else if (testResponse.status >= 500) {
-          logger.warn(
-            `Vision test for ${cacheKey}: ${testResponse.status} ${testResponse.statusText}, not caching`,
-          );
-          response.status(502).json({
-            error: `Unable to verify vision support for model — upstream returned ${testResponse.status}`,
-            model,
-            provider,
-          });
-        } else {
-          ModelCapabilitiesCache.set(cacheKey, false);
-          response.json({ model, provider, supportsVision: false });
         }
       } catch (error) {
         logger.error(`Vision test error for ${cacheKey}:`, error);