fix(hermes): avoid Anthropic SSE for compatible endpoints

docs/inference/switch-inference-providers.mdx

@@ -111,8 +111,8 @@ For OpenClaw, `inference set` syncs the provider API family and primary model re
 For Hermes, `inference set` writes `model.api_mode: anthropic_messages` for Anthropic Messages routes, `model.api_mode: codex_responses` for OpenAI Responses routes, and removes `api_mode` for OpenAI-style chat-completions routes.
 Hermes also keeps `model.api_key` on the OpenShell proxy placeholder so dashboard and API sessions continue to authenticate through the gateway after a route change.
 
-Amazon Bedrock Runtime routes created through `compatible-anthropic-endpoint` are the exception.
-When you switch within the same Bedrock Runtime compatible provider, NemoClaw keeps the route OpenAI-compatible and does not set Hermes to Anthropic Messages mode.
+Hermes only uses Anthropic Messages mode for routes that NemoClaw resolves as true Anthropic Messages providers, such as `anthropic-prod`.
+For `compatible-anthropic-endpoint`, NemoClaw keeps the Hermes route OpenAI-compatible through the managed `/v1` proxy and does not set Hermes to Anthropic Messages mode.
 
 #### Switching from Responses API to Chat Completions
 

src/lib/actions/inference-route-api.test.ts

@@ -192,10 +192,30 @@ describe("resolveRuntimeInferenceApi", () => {
     expect(
       resolve(
         { model: { api_mode: "anthropic_messages" } },
-        { agentName: "hermes", session: null },
+        {
+          agentName: "hermes",
+          currentProvider: "anthropic-prod",
+          provider: "anthropic-prod",
+          session: null,
+        },
       ),
     ).toBe("anthropic-messages");
   });
+
+  it("keeps Hermes compatible Anthropic endpoint switches off Anthropic SSE streaming (#6289)", () => {
+    expect(
+      resolve(
+        { model: { api_mode: "anthropic_messages" } },
+        {
+          agentName: "hermes",
+          session: session({
+            agent: "hermes",
+            preferredInferenceApi: "anthropic-messages",
+          }),
+        },
+      ),
+    ).toBe("openai-completions");
+  });
 });
 
 describe("hermesApiMode", () => {

src/lib/actions/inference-route-api.ts

@@ -109,6 +109,9 @@ export function resolveRuntimeInferenceApi(options: {
 }): InferenceApi | null {
   const { agentName, config, currentProvider, provider, sandboxName, session } = options;
   if (provider === "anthropic-prod") return "anthropic-messages";
+  if (agentName === "hermes" && provider === "compatible-anthropic-endpoint") {
+    return "openai-completions";
+  }
 
   const sameProvider = currentProvider === provider;
   const sessionApi = sameProvider ? sessionRouteApi(session, sandboxName, provider) : null;

src/lib/actions/inference-set-hermes-run.test.ts

@@ -104,7 +104,7 @@ describe("runInferenceSet Hermes routing", () => {
     });
   });
 
-  it("syncs Hermes compatible Anthropic switches to Anthropic Messages when changing provider families", async () => {
+  it("keeps Hermes compatible Anthropic switches on the managed OpenAI-compatible route (#6289)", async () => {
     const config: ConfigObject = {
       model: {
         default: "openai/gpt-5.4-mini",
@@ -146,9 +146,8 @@ describe("runInferenceSet Hermes routing", () => {
     expect(config.model).toEqual({
       default: "claude-sonnet-proxy",
       provider: "custom",
-      base_url: "https://inference.local",
+      base_url: "https://inference.local/v1",
       api_key: HERMES_PROXY_API_KEY_PLACEHOLDER,
-      api_mode: "anthropic_messages",
     });
     // The upstream annotation must track the selected provider together with
     // the API-family field, so the two cannot drift apart on later switches.
@@ -163,17 +162,17 @@ describe("runInferenceSet Hermes routing", () => {
         model: "claude-sonnet-proxy",
         endpointUrl: "https://anthropic-compatible.example/v1",
         credentialEnv: "COMPATIBLE_ANTHROPIC_API_KEY",
-        preferredInferenceApi: "anthropic-messages",
+        preferredInferenceApi: "openai-completions",
       }),
     ]);
     expect(deps.getSession()).toMatchObject({
       provider: "compatible-anthropic-endpoint",
       model: "claude-sonnet-proxy",
-      preferredInferenceApi: "anthropic-messages",
+      preferredInferenceApi: "openai-completions",
     });
     expect(result).toMatchObject({
-      providerKey: "anthropic",
-      primaryModelRef: "anthropic/claude-sonnet-proxy",
+      providerKey: "inference",
+      primaryModelRef: "inference/claude-sonnet-proxy",
     });
   });