feat: add databricks fable 5.1 model capabilities (#7213)

## Summary - add the canonical public Databricks `Claude Fable 5.1` model record - match the existing Fable family contract: adaptive thinking, `low|medium|high|xhigh|max`, default `high`, Anthropic Messages, and no normalization - add generated Rust/TypeScript corpus coverage and canonical Global Defaults label/persistence coverage - remove deployment-specific catalog references from the repository ## Scope This is metadata and regression coverage only. It does not change model-capability resolution or production frontend logic. ## Validation - `cargo fmt --check` - `node --test desktop/src/features/agents/ui/modelCapabilitiesCorpus.test.mjs` - `cargo test -p buzz-agent` - `pnpm build:e2e` - `pnpm exec playwright test --project=smoke --grep 'defaults render the Fable 5.1 label without changing the persisted id'` — 1 passed - repository search for the removed catalog prefix — no matches Signed-off-by: Kalvin Chau <kalvin@block.xyz> Co-authored-by: am <6e30cd56c30e030cd31bb0939b94a7c257c9a09d5ba2d92cf2735da45629f248@buzz.block.builderlab.xyz>
2026-10-27 12:40:30 +00:00 · 2026-09-01 16:06:17 -07:00
parent d4420eb47f
commit 560fea7b4f
5 changed files with 103 additions and 32 deletions
--- a/crates/buzz-agent/src/model_capabilities.rs
+++ b/crates/buzz-agent/src/model_capabilities.rs
@@ -634,6 +634,7 @@ mod tests {
    Q::Vector { id: "resolver-exact-raw-id-probe", provider: "databricks_v2", raw_model_id: "databricks-gpt-5-4-mini", note: Some("Probes a raw id that has an exact record.") },
    Q::Vector { id: "dbv2-claude-fable-5-exact-record-probe", provider: "databricks_v2", raw_model_id: "databricks-claude-fable-5", note: Some("Probes the canonical Databricks Fable 5 endpoint record.") },
    Q::Vector { id: "dbv2-goose-claude-fable-5-alias-probe", provider: "databricks_v2", raw_model_id: "goose-claude-fable-5", note: Some("Probes a prefixed alias of the Databricks Fable 5 endpoint.") },
+    Q::Vector { id: "dbv2-claude-fable-5-1-exact-record-probe", provider: "databricks_v2", raw_model_id: "databricks-claude-fable-5-1", note: Some("Probes the canonical Databricks Fable 5.1 endpoint record.") },
    Q::Vector { id: "dbv2-claude-opus-4-8-exact-record-probe", provider: "databricks_v2", raw_model_id: "databricks-claude-opus-4-8", note: Some("Probes the canonical Databricks Opus 4.8 endpoint record.") },
    Q::Vector { id: "dbv2-goose-claude-opus-4-8-alias-probe", provider: "databricks_v2", raw_model_id: "goose-claude-opus-4-8", note: Some("Probes a prefixed alias of the Databricks Opus 4.8 endpoint.") },
    Q::Vector { id: "dbv2-claude-opus-5-exact-record-probe", provider: "databricks_v2", raw_model_id: "databricks-claude-opus-5", note: Some("Probes the canonical Databricks Opus 5 endpoint record.") },
@@ -743,7 +744,7 @@ mod tests {
    Q::Vector { id: "dbv2-inkling-exact-record-probe", provider: "databricks_v2", raw_model_id: "databricks-inkling", note: Some("Probes the Inkling endpoint record and label.") },
    Q::Vector { id: "dbv2-uc-fqn-gemini-3-5-flash-strip-probe", provider: "databricks_v2", raw_model_id: "system.ai.gemini-3-5-flash", note: Some("Probes strip parity on a system.ai. UC FQN carrying the gemini- token (resolve carries no label; the alias label path is unit-tested).") },
    Q::Vector { id: "dbv2-uc-fqn-meta-llama-strip-probe", provider: "databricks_v2", raw_model_id: "system.ai.meta-llama-3-3-70b-instruct", note: Some("Probes strip parity on a UC FQN where the llama- token strips through meta-.") },
-    Q::Vector { id: "dbv2-uc-goose-deepseek-strip-probe", provider: "databricks_v2", raw_model_id: "data_workflow_tools.goose.goose-deepseek-v4-pro-0813", note: Some("Probes strip parity on a goose- prefixed UC FQN carrying the deepseek- token.") },
+    Q::Vector { id: "dbv2-uc-fqn-deepseek-strip-probe", provider: "databricks_v2", raw_model_id: "system.ai.deepseek-v4-pro-0813", note: Some("Probes strip parity on a UC FQN carrying the deepseek- token.") },
    Q::Vector { id: "dbv2-uc-fqn-inkling-strip-probe", provider: "databricks_v2", raw_model_id: "system.ai.inkling", note: Some("Probes strip parity on a UC FQN carrying the bare inkling token.") },
    Q::Section { group: "Label/capability token isolation probes (#6955 review pass 1)", note: Some("Pins that label_family_tokens (the UC-humanization superset) never leaks into capability resolve(): capability stripping still uses only claude-/gpt-/kimi-, so a label token appearing before a gpt- marker must NOT displace the gpt-5-pro exact profile.") },
    Q::Vector { id: "isolation-openai-gemini-gpt-5-pro-probe", provider: "openai", raw_model_id: "tenant-gemini-gpt-5-pro", note: Some("The gemini- label token must not strip here; capability resolve keeps the gpt-5-pro high-only profile.") },
@@ -843,7 +844,7 @@ mod tests {
    }

    #[test]
-    fn corpus_has_exactly_139_executable_vectors() {
+    fn corpus_has_exactly_140_executable_vectors() {
        // Locks the vector count so a silent INPUTS edit can't quietly drop
        // coverage; must equal the gate in the TS harness
        // (modelCapabilitiesCorpus.test.mjs).
@@ -852,7 +853,7 @@ mod tests {
            .filter(|q| matches!(q, Q::Vector { .. }))
            .count();
        assert_eq!(
-            vectors, 139,
+            vectors, 140,
            "corpus executable-vector count changed; update this gate deliberately"
        );
    }
@@ -873,7 +874,7 @@ mod tests {

    #[test]
    fn databricks_v2_fqn_uses_neutral_concrete_unknown_capabilities() {
-        let fqn = resolve("databricks_v2", "data_workflow_tools.goose.goose-kimi-k3");
+        let fqn = resolve("databricks_v2", "system.ai.kimi-k3");
        let fallback = resolve("databricks_v2", "some-unknown-xyz");
        assert_eq!(fqn.thinking_mode, fallback.thinking_mode);
        assert_eq!(fqn.supported_efforts, fallback.supported_efforts);
@@ -1056,16 +1057,10 @@ mod tests {
            ("system.ai.qwen35-122b-a10b", "Qwen3.5 122B A10B"),
            ("system.ai.gemma-3-12b", "Gemma 3 12B"),
            ("system.ai.inkling", "Inkling"),
-            (
-                "data_workflow_tools.goose.goose-deepseek-v4-flash-0731",
-                "DeepSeek V4 Flash",
-            ),
-            ("data_workflow_tools.goose.goose-glm-5-3", "GLM-5.3"),
-            (
-                "data_workflow_tools.goose.goose-glm-5-3-flash",
-                "GLM-5.3 Flash",
-            ),
-            ("data_workflow_tools.goose.goose-grok-4-6", "Grok 4.6"),
+            ("system.ai.deepseek-v4-flash-0731", "DeepSeek V4 Flash"),
+            ("system.ai.glm-5-3", "GLM-5.3"),
+            ("system.ai.glm-5-3-flash", "GLM-5.3 Flash"),
+            ("system.ai.grok-4-6", "Grok 4.6"),
        ] {
            assert_eq!(databricks_registry_label(fqn), Some(label), "fqn={fqn}");
        }
--- a/desktop/src/features/agents/ui/modelCapabilitiesCorpus.test.mjs
+++ b/desktop/src/features/agents/ui/modelCapabilitiesCorpus.test.mjs
@@ -25,10 +25,10 @@ const corpus = JSON.parse(readFileSync(fileURLToPath(corpusUrl), "utf8"));
 // (`_group`) are skipped. Mirrors the Rust corpus filter.
 const executable = corpus.filter((entry) => entry.expect != null);

-test("corpus has exactly 139 executable vectors", () => {
+test("corpus has exactly 140 executable vectors", () => {
  // Locks the vector count so a silent corpus edit can't quietly drop coverage;
  // must equal the gate in the Rust suite (model_capabilities.rs).
-  assert.equal(executable.length, 139);
+  assert.equal(executable.length, 140);
 });

 test("registry label aliases refuse an unprefixed query", () => {
@@ -45,7 +45,7 @@ test("registry label aliases refuse an unprefixed query", () => {
  );
 });

-test("UC model-family FQNs and goose- aliases humanize onto their base records", () => {
+test("UC model-family FQNs humanize onto their base records", () => {
  // #6918 follow-up: the shared UC-FQN (`system.ai.…`) and goose- alias forms
  // must resolve onto the same base databricks_v2 records via the new family
  // tokens. Mirrors the Rust `test_databricks_registry_label_lookup` coverage.
@@ -64,13 +64,10 @@ test("UC model-family FQNs and goose- aliases humanize onto their base records",
    ["system.ai.qwen35-122b-a10b", "Qwen3.5 122B A10B"],
    ["system.ai.gemma-3-12b", "Gemma 3 12B"],
    ["system.ai.inkling", "Inkling"],
-    [
-      "data_workflow_tools.goose.goose-deepseek-v4-flash-0731",
-      "DeepSeek V4 Flash",
-    ],
-    ["data_workflow_tools.goose.goose-glm-5-3", "GLM-5.3"],
-    ["data_workflow_tools.goose.goose-glm-5-3-flash", "GLM-5.3 Flash"],
-    ["data_workflow_tools.goose.goose-grok-4-6", "Grok 4.6"],
+    ["system.ai.deepseek-v4-flash-0731", "DeepSeek V4 Flash"],
+    ["system.ai.glm-5-3", "GLM-5.3"],
+    ["system.ai.glm-5-3-flash", "GLM-5.3 Flash"],
+    ["system.ai.grok-4-6", "Grok 4.6"],
  ];
  for (const [fqn, label] of cases) {
    assert.equal(databricksRegistryLabel(fqn), label, `fqn=${fqn}`);
@@ -97,10 +94,7 @@ test("registry label aliases refuse ambiguous stripped record keys", () => {
 });

 test("Unity Catalog FQNs use neutral concrete-unknown capabilities", () => {
-  const fqn = resolveModelCapabilities(
-    "databricks_v2",
-    "data_workflow_tools.goose.goose-kimi-k3",
-  );
+  const fqn = resolveModelCapabilities("databricks_v2", "system.ai.kimi-k3");
  const fallback = resolveModelCapabilities(
    "databricks_v2",
    "some-unknown-xyz",
--- a/desktop/tests/e2e/global-agent-config-screenshots.spec.ts
+++ b/desktop/tests/e2e/global-agent-config-screenshots.spec.ts
@@ -251,7 +251,7 @@ test.describe("global agent config screenshots", () => {
  test("defaults render Databricks model labels without changing persisted ids", async ({
    page,
  }) => {
-    const modelId = "data_workflow_tools.goose.goose-glm-5-3";
+    const modelId = "system.ai.glm-5-3";
    await installMockBridge(page, {
      globalAgentConfig: {
        preferred_runtime: "goose",
@@ -291,6 +291,50 @@ test.describe("global agent config screenshots", () => {
    expect(persisted).toMatchObject({ model: modelId });
  });

+  test("defaults render the Fable 5.1 label without changing the persisted id", async ({
+    page,
+  }) => {
+    const modelId = "databricks-claude-fable-5-1";
+    await installMockBridge(page, {
+      globalAgentConfig: {
+        preferred_runtime: "goose",
+        provider: "databricks_v2",
+        model: modelId,
+        env_vars: {},
+      },
+      discoverAgentModels: {
+        models: [{ id: modelId, name: modelId }],
+        supportsSwitching: true,
+        selectedModel: modelId,
+      },
+      runtimeFileConfigs: {
+        goose: {
+          provider: "databricks_v2",
+          model: modelId,
+          satisfiedEnvKeys: ["DATABRICKS_HOST"],
+        },
+      },
+    });
+
+    await openAiDefaultsSettings(page);
+
+    const model = page.getByTestId("global-agent-model");
+    await expect(model).toHaveText("Claude Fable 5.1");
+    await expect(model).toHaveAttribute("data-value", modelId);
+
+    const persisted = await page.evaluate(async () =>
+      (
+        window as typeof window & {
+          __BUZZ_E2E_INVOKE_MOCK_COMMAND__?: (
+            command: string,
+            payload: unknown,
+          ) => Promise<unknown>;
+        }
+      ).__BUZZ_E2E_INVOKE_MOCK_COMMAND__?.("get_global_agent_config", null),
+    );
+    expect(persisted).toMatchObject({ model: modelId });
+  });
+
  test("defaults honor credentials set in the harness config file", async ({
    page,
  }) => {
--- a/scripts/model-capabilities.json
+++ b/scripts/model-capabilities.json
@@ -487,6 +487,24 @@
      "_provenance": "all axes materialized from family:anthropic-adaptive-xhigh-fable-5",
      "_source": "models.dev anthropic catalog label; Databricks workspace endpoint is absent from its provider catalog"
    },
+    {
+      "provider": "databricks_v2",
+      "raw_model_id": "databricks-claude-fable-5-1",
+      "registry_label": "Claude Fable 5.1",
+      "thinking_mode": "adaptive",
+      "supported_efforts": [
+        "low",
+        "medium",
+        "high",
+        "xhigh",
+        "max"
+      ],
+      "default_effort": "high",
+      "databricks_v2_wire_route": "anthropic-messages",
+      "normalization_policy": "none",
+      "_provenance": "all axes materialized from family:anthropic-adaptive-xhigh-fable-5",
+      "_source": "models.dev anthropic catalog label; Databricks workspace endpoint is absent from its provider catalog"
+    },
    {
      "provider": "databricks_v2",
      "raw_model_id": "databricks-claude-opus-4-8",
--- a/scripts/normative-corpus.json
+++ b/scripts/normative-corpus.json
@@ -759,6 +759,26 @@
      "registry_label": null
    }
  },
+  {
+    "id": "dbv2-claude-fable-5-1-exact-record-probe",
+    "provider": "databricks_v2",
+    "raw_model_id": "databricks-claude-fable-5-1",
+    "_note": "Probes the canonical Databricks Fable 5.1 endpoint record.",
+    "expect": {
+      "thinking_mode": "adaptive",
+      "supported_efforts": [
+        "low",
+        "medium",
+        "high",
+        "xhigh",
+        "max"
+      ],
+      "default_effort": "high",
+      "databricks_v2_wire_route": "anthropic-messages",
+      "normalization_policy": "none",
+      "registry_label": "Claude Fable 5.1"
+    }
+  },
  {
    "id": "dbv2-claude-opus-4-8-exact-record-probe",
    "provider": "databricks_v2",
@@ -2708,10 +2728,10 @@
    }
  },
  {
-    "id": "dbv2-uc-goose-deepseek-strip-probe",
+    "id": "dbv2-uc-fqn-deepseek-strip-probe",
    "provider": "databricks_v2",
-    "raw_model_id": "data_workflow_tools.goose.goose-deepseek-v4-pro-0813",
-    "_note": "Probes strip parity on a goose- prefixed UC FQN carrying the deepseek- token.",
+    "raw_model_id": "system.ai.deepseek-v4-pro-0813",
+    "_note": "Probes strip parity on a UC FQN carrying the deepseek- token.",
    "expect": {
      "thinking_mode": "none",
      "supported_efforts": [