fix(vision): honor modelCapabilities text-only on native rows (#4501)

src/vision/eligibility.ts

@@ -156,9 +156,10 @@ function modelAcceptsImageInputWithCache(
 ): boolean | undefined {
   if (candidate.native === true || (candidate.provider === "openai" && SUPPORTED_NATIVE_OPENAI_SLUGS.has(candidate.id))) {
     const nativeProvider = enrichedProviderForVision(config, candidate.provider, cache);
-    if (nativeProvider && isModelVisionSidecarConsumer({
-      noVisionModels: nativeProvider.noVisionModels, modelInputModalities: nativeProvider.modelInputModalities,
-    }, candidate.id)) return false;
+    if (nativeProvider && isModelVisionSidecarConsumer(nativeProvider, candidate.id)) return false;
+    const declared = Object.hasOwn(nativeProvider?.modelCapabilities ?? {}, candidate.id)
+      ? nativeProvider?.modelCapabilities?.[candidate.id]?.inputModalities : undefined;
+    if (declared !== undefined) return declared.includes("image");
     return advertisesImageInput(nativeInputModalities(candidate.id)) ?? true;
   }
   if (isVisionSidecarConsumerWithCache(config, candidate.provider, candidate.id, cache)) return false;

tests/vision/vision-eligibility.test.ts

@@ -227,6 +227,42 @@ describe("vision eligibility core", () => {
     })).toBe(false);
   });
 
+  test("15. native model declared text-only via modelCapabilities is disqualified from vision describer eligibility (#4501)", () => {
+    const config = configWithProviders({
+      openai: {
+        adapter: "openai-responses",
+        baseUrl: "https://chatgpt.com/backend-api/codex",
+        modelCapabilities: {
+          "gpt-5.4-mini": { inputModalities: ["text"] },
+        },
+      },
+    });
+    const candidate: VisionCandidateModel = {
+      provider: "openai",
+      id: "gpt-5.4-mini",
+    };
+    expect(modelAcceptsImageInput(config, candidate)).toBe(false);
+    expect(isVisionEligibleModel(config, candidate)).toBe(false);
+  });
+
+  test("16. native model baseline declared text-only via modelCapabilities drops the baseline option (#4501)", () => {
+    const config = configWithProviders({
+      openai: {
+        adapter: "openai-responses",
+        baseUrl: "https://chatgpt.com/backend-api/codex",
+        modelCapabilities: {
+          [BASELINE_VISION_MODELS.openai]: { inputModalities: ["text"] },
+        },
+      },
+      anthropic: {
+        adapter: "anthropic",
+        baseUrl: "https://api.anthropic.com",
+      },
+    });
+    const options = visionEligibleModelOptions(config, [], ["openai", "anthropic"], "anthropic");
+    expect(options.map(o => o.value)).toEqual([BASELINE_VISION_MODELS.anthropic]);
+  });
+
   test("5. openai baseline is present when that side is enabled", () => {
     const options = visionEligibleModelOptions(emptyConfig, [], ["openai"]);
     expect(options.map((o) => o.value)).toEqual([BASELINE_VISION_MODELS.openai]);