diff --git a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/index.ts b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/index.ts
index e48428c9c732..42d2d895fea0 100644
--- a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/index.ts
+++ b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/index.ts
@@ -34,6 +34,14 @@ export default Sentry.withSentry(
         return new Response(JSON.stringify(result));
       }
 
+      if (url.pathname === '/evaluate') {
+        const result = await ai.run('typesafe/jev', {
+          state: 'Help! My payouts have been failing for 3 days.',
+          questions: { is_urgent: { type: 'noul', instructions: 'Does this convey urgency?' } },
+        });
+        return new Response(JSON.stringify(result));
+      }
+
       if (url.pathname === '/stream') {
         const stream = (await ai.run('@cf/meta/llama-3.1-8b-instruct', {
           messages: [{ role: 'user', content: 'What is the capital of France?' }],
diff --git a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/mocks.ts b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/mocks.ts
index bce096d4ae75..e0a26d02bf86 100644
--- a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/mocks.ts
+++ b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/mocks.ts
@@ -23,6 +23,17 @@ export class MockAi {
       throw error;
     }
 
+    if (model === 'typesafe/jev') {
+      return {
+        state: 'Completed',
+        result: {
+          model: 'jev-1.13.0',
+          answers: { is_urgent: { type: 'noul', noul: 0.97 } },
+          usage: { input_tokens: 426, output_tokens: 73 },
+        },
+      };
+    }
+
     if (inputs?.stream === true) {
       return createSseStream([
         '{"response":"The capital "}',
diff --git a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/test.ts b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/test.ts
index 726f91a5e467..4eaf940ef9c7 100644
--- a/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/test.ts
+++ b/dev-packages/cloudflare-integration-tests/suites/tracing/workers-ai/test.ts
@@ -5,6 +5,7 @@ import {
   GEN_AI_REQUEST_MAX_TOKENS,
   GEN_AI_REQUEST_MODEL,
   GEN_AI_REQUEST_TEMPERATURE,
+  GEN_AI_RESPONSE_MODEL,
   GEN_AI_RESPONSE_STREAMING,
   GEN_AI_RESPONSE_TEXT,
   GEN_AI_USAGE_INPUT_TOKENS,
@@ -132,6 +133,52 @@ it('traces a streaming Workers AI text generation request', async ({ signal }) =
   await runner.completed();
 });
 
+it('traces a TypeSafe Jev evaluation like the TypeSafe integration', async ({ signal }) => {
+  const runner = createRunner(__dirname)
+    .ignore('event')
+    .expect(envelope => {
+      const spans = getSpansFromEnvelope(envelope);
+      const segmentSpan = spans.find(span => span.is_segment);
+
+      const genAiSpans = spans.filter(span => getSpanOp(span)?.startsWith('gen_ai.'));
+      expect(genAiSpans).toHaveLength(1);
+
+      expect(genAiSpans[0]).toEqual(
+        expect.objectContaining({
+          name: 'evaluate typesafe/jev',
+          status: 'ok',
+          is_segment: false,
+          attributes: {
+            'sentry.origin': { value: 'auto.ai.cloudflare.workers_ai', type: 'string' },
+            'sentry.op': { value: 'gen_ai.evaluate', type: 'string' },
+            [GEN_AI_PROVIDER_NAME]: { value: 'cloudflare.workers_ai', type: 'string' },
+            [GEN_AI_OPERATION_NAME]: { value: 'evaluate', type: 'string' },
+            [GEN_AI_REQUEST_MODEL]: { value: 'typesafe/jev', type: 'string' },
+            [GEN_AI_RESPONSE_MODEL]: { value: 'jev-1.13.0', type: 'string' },
+            [GEN_AI_USAGE_INPUT_TOKENS]: { value: 426, type: 'integer' },
+            [GEN_AI_USAGE_OUTPUT_TOKENS]: { value: 73, type: 'integer' },
+            [GEN_AI_USAGE_TOTAL_TOKENS]: { value: 499, type: 'integer' },
+            // collect only output messages
+            [GEN_AI_OUTPUT_MESSAGES]: {
+              type: 'string',
+              value: '[{"type":"evaluation","answers":{"is_urgent":{"type":"noul","noul":0.97}}}]',
+            },
+            'sentry.is_localhost': { value: true, type: 'boolean' },
+            [SENTRY_TRACE_LIFECYCLE]: { value: 'stream', type: 'string' },
+            [SENTRY_SEGMENT_NAME]: { value: segmentSpan!.name, type: 'string' },
+            [SENTRY_SEGMENT_ID]: { value: segmentSpan!.span_id, type: 'string' },
+            [SENTRY_SDK_NAME]: { value: 'sentry.javascript.cloudflare', type: 'string' },
+            [SENTRY_SDK_VERSION]: { value: SDK_VERSION, type: 'string' },
+            [SEMANTIC_ATTRIBUTE_SENTRY_ENVIRONMENT]: { value: 'production', type: 'string' },
+          },
+        }),
+      );
+    })
+    .start(signal);
+  await runner.makeRequest('get', '/evaluate');
+  await runner.completed();
+});
+
 // The Workers AI integration deliberately does not call `captureException` itself.
 // When a `run` call fails, the error must bubble up out of the fetch handler and be
 // reported by the top-level Cloudflare instrumentation instead — so it shows up in
diff --git a/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/src/index.ts b/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/src/index.ts
index 324c9161cf69..87ae6c9cff41 100644
--- a/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/src/index.ts
+++ b/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/src/index.ts
@@ -31,6 +31,13 @@ export default {
         await env.AI.run('@cf/meta/llama-3.2-1b-instruct', { prompt: 'Say hi', max_tokens: 5 });
         return Response.json({ traceId: spanContext?.traceId });
       }
+      case '/test-workers-ai-jev': {
+        await env.AI.run('typesafe/jev', {
+          state: 'Help! My payouts have been failing for 3 days.',
+          questions: { is_urgent: { type: 'noul', instructions: 'Does this convey urgency?' } },
+        });
+        return Response.json({ traceId: spanContext?.traceId });
+      }
       case '/test-span':
         return Response.json({ spanId: spanContext?.spanId, traceId: spanContext?.traceId });
       case '/test-workflow-sleep': {
diff --git a/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/tests/send-to-sentry.test.ts b/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/tests/send-to-sentry.test.ts
index 2397db37089f..b391f2f8244b 100644
--- a/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/tests/send-to-sentry.test.ts
+++ b/dev-packages/e2e-tests/test-applications/cloudflare-workers-send-to-sentry/tests/send-to-sentry.test.ts
@@ -93,3 +93,24 @@ test('Sends a Workers AI gen_ai span to Sentry', async () => {
       'gen_ai.output.messages': expect.stringContaining('"role":"assistant"'),
     });
 });
+
+test('Sends a Workers AI gen_ai.evaluate span for a TypeSafe Jev call to Sentry', async () => {
+  const { traceId }: { traceId: string } = JSON.parse(await fetchFromWorker(`${workerUrl}/test-workers-ai-jev`, 200));
+
+  console.log(`Polling for gen_ai.evaluate span: sentry trace view ${traceTarget(traceId)}`);
+
+  let spanId: string | undefined;
+  await expect
+    .poll(() => (spanId = findSpanInTrace(traceId, 'gen_ai.evaluate')?.event_id), EVENT_POLLING_OPTIONS)
+    .toBeDefined();
+
+  await expect
+    .poll(() => fetchSpanAttributes(traceId, spanId!), EVENT_POLLING_OPTIONS)
+    .toMatchObject({
+      'gen_ai.operation.name': 'evaluate',
+      'gen_ai.request.model': 'typesafe/jev',
+      'gen_ai.response.model': expect.stringMatching(/^jev-/),
+      'gen_ai.input.messages': expect.stringContaining('My payouts have been failing'),
+      'gen_ai.output.messages': expect.stringContaining('"is_urgent"'),
+    });
+});
diff --git a/packages/server-utils/src/ai/typesafe/index.ts b/packages/server-utils/src/ai/typesafe/index.ts
index b8fe48878024..328853cf5b02 100644
--- a/packages/server-utils/src/ai/typesafe/index.ts
+++ b/packages/server-utils/src/ai/typesafe/index.ts
@@ -49,17 +49,19 @@ function getRequestAttributes(
     [GEN_AI_OPERATION_NAME]: 'evaluate',
     [GEN_AI_PROVIDER_NAME]: TYPESAFE_PROVIDER_NAME,
     ...(model ? { [GEN_AI_REQUEST_MODEL]: model } : {}),
-    ...(recordInputs
-      ? {
-          [GEN_AI_INPUT_MESSAGES]: stringify([
-            { type: 'evaluation', state: request.state, questions: request.questions },
-          ]),
-        }
-      : {}),
+    ...(recordInputs ? { [GEN_AI_INPUT_MESSAGES]: getEvaluationInputMessages(request) } : {}),
   };
 }
 
-/** Add the response model, token usage and (optionally) the answers of a `systemOne` result. */
+/** Serialize the `state` and `questions` of an evaluation request. Also used for TypeSafe models on Workers AI. */
+export function getEvaluationInputMessages(request: Record<string, unknown>): string | undefined {
+  return stringify([{ type: 'evaluation', state: request.state, questions: request.questions }]);
+}
+
+/**
+ * Add the response model, token usage and (optionally) the answers of a `systemOne` result.
+ * Also used for TypeSafe models on Workers AI.
+ */
 export function addResponseAttributes(span: Span, result: unknown, recordOutputs: boolean): void {
   if (!isObjectLike(result)) {
     return;
diff --git a/packages/server-utils/src/ai/workers-ai/index.ts b/packages/server-utils/src/ai/workers-ai/index.ts
index c02222cc1877..eb3a7c5407cb 100644
--- a/packages/server-utils/src/ai/workers-ai/index.ts
+++ b/packages/server-utils/src/ai/workers-ai/index.ts
@@ -89,7 +89,7 @@ function instrumentRun(
           }
 
           // The model did not actually return a stream — finalize the span eagerly.
-          addResponseAttributes(span, result, options.recordOutputs);
+          addResponseAttributes(span, result, options.recordOutputs, operationName);
           span.end();
           return result;
         }, handleError);
@@ -105,7 +105,7 @@ function instrumentRun(
 
       return originalResult.then(result => {
         if (!returnsRawResponse) {
-          addResponseAttributes(span, result, options.recordOutputs);
+          addResponseAttributes(span, result, options.recordOutputs, operationName);
         }
         return result;
       });
diff --git a/packages/server-utils/src/ai/workers-ai/utils.ts b/packages/server-utils/src/ai/workers-ai/utils.ts
index 33908290be72..67645a456988 100644
--- a/packages/server-utils/src/ai/workers-ai/utils.ts
+++ b/packages/server-utils/src/ai/workers-ai/utils.ts
@@ -16,10 +16,16 @@ import {
   GEN_AI_SYSTEM_INSTRUCTIONS,
 } from '@sentry/conventions/attributes';
 import { GEN_AI_CHAT, GEN_AI_EMBEDDINGS } from '@sentry/conventions/op';
-import { SEMANTIC_ATTRIBUTE_SENTRY_ORIGIN, stringify } from '@sentry/core';
+import { isObjectLike, SEMANTIC_ATTRIBUTE_SENTRY_ORIGIN, stringify } from '@sentry/core';
 import type { Span, SpanAttributeValue } from '@sentry/core';
 import { GEN_AI_REQUEST_STREAM_ATTRIBUTE } from '../core/gen-ai-attributes';
-import { extractSystemInstructions, setOutputMessagesAttribute, setTokenUsageAttributes } from '../core/utils';
+import {
+  extractSystemInstructions,
+  getGenAiSpanOp,
+  setOutputMessagesAttribute,
+  setTokenUsageAttributes,
+} from '../core/utils';
+import { addResponseAttributes as addEvaluateResponseAttributes, getEvaluationInputMessages } from '../typesafe';
 // Re-exported so `workers-ai/streaming.ts` keeps importing it from this module.
 export { setOutputMessagesAttribute };
 import { WORKERS_AI_ORIGIN, WORKERS_AI_PROVIDER_NAME } from './constants';
@@ -29,15 +35,20 @@ import type { WorkersAiInput, WorkersAiOutput } from './types';
  * Determine the gen_ai operation name from the inputs passed to `AI.run`.
  * Workers AI exposes a single `run` method, so we infer the operation from the input shape.
  */
-export type WorkersAiOperationName = 'chat' | 'embeddings';
+export type WorkersAiOperationName = 'chat' | 'embeddings' | 'evaluate';
 
 export const WORKERS_AI_OPERATION_SPAN_OPS: Record<WorkersAiOperationName, string> = {
   chat: GEN_AI_CHAT,
   embeddings: GEN_AI_EMBEDDINGS,
+  evaluate: getGenAiSpanOp('evaluate'),
 };
 
 export function getOperationName(inputs: unknown): WorkersAiOperationName {
   if (inputs && typeof inputs === 'object') {
+    // TypeSafe evaluation models (e.g. `typesafe/jev`)
+    if ('state' in inputs && 'questions' in inputs) {
+      return 'evaluate';
+    }
     if ('messages' in inputs || 'prompt' in inputs) {
       return 'chat';
     }
@@ -102,6 +113,11 @@ export function addRequestAttributes(span: Span, inputs: unknown, operationName:
   }
   const params = inputs as WorkersAiInput;
 
+  if (operationName === 'evaluate') {
+    span.setAttribute(GEN_AI_INPUT_MESSAGES, getEvaluationInputMessages(inputs as Record<string, unknown>));
+    return;
+  }
+
   // Store embeddings input on a separate attribute
   if (operationName === 'embeddings') {
     const text = params.text;
@@ -132,7 +148,19 @@ export function addRequestAttributes(span: Span, inputs: unknown, operationName:
 /**
  * Record the response attributes (token usage, response text, tool calls) on the span.
  */
-export function addResponseAttributes(span: Span, result: unknown, recordOutputs: boolean): void {
+export function addResponseAttributes(
+  span: Span,
+  result: unknown,
+  recordOutputs: boolean,
+  operationName?: WorkersAiOperationName,
+): void {
+  if (operationName === 'evaluate') {
+    // The binding wraps the TypeSafe response as `{ state, result }`, but the docs show it unwrapped.
+    const body = isObjectLike(result) && isObjectLike(result.result) ? result.result : result;
+    addEvaluateResponseAttributes(span, body, recordOutputs);
+    return;
+  }
+
   if (
     !result ||
     typeof result !== 'object' ||
diff --git a/packages/server-utils/test/ai/lib/tracing/workers-ai.test.ts b/packages/server-utils/test/ai/lib/tracing/workers-ai.test.ts
index 4baced963ec2..bc282e0348bc 100644
--- a/packages/server-utils/test/ai/lib/tracing/workers-ai.test.ts
+++ b/packages/server-utils/test/ai/lib/tracing/workers-ai.test.ts
@@ -5,8 +5,12 @@ import {
   GEN_AI_OUTPUT_MESSAGES,
   GEN_AI_PROVIDER_NAME,
   GEN_AI_REQUEST_MODEL,
+  GEN_AI_RESPONSE_MODEL,
   GEN_AI_RESPONSE_TEXT,
   GEN_AI_SYSTEM_INSTRUCTIONS,
+  GEN_AI_USAGE_INPUT_TOKENS,
+  GEN_AI_USAGE_OUTPUT_TOKENS,
+  GEN_AI_USAGE_TOTAL_TOKENS,
 } from '@sentry/conventions/attributes';
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 import {
@@ -160,6 +164,45 @@ describe('instrumentWorkersAiClient', () => {
     });
   });
 
+  it('records TypeSafe models as evaluate spans, like the TypeSafe integration', async () => {
+    const client = new TestClient(
+      getDefaultTestClientOptions({ dsn: 'https://public@dsn.ingest.sentry.io/1337', tracesSampleRate: 1 }),
+    );
+    setCurrentClient(client);
+    client.init();
+    const endedSpans: Span[] = [];
+    client.on('spanEnd', span => endedSpans.push(span));
+
+    const questions = { is_urgent: { type: 'noul', instructions: 'Does this convey urgency?' } };
+    const answers = { is_urgent: { type: 'noul', noul: 0.97 } };
+    const ai = {
+      run: vi.fn().mockResolvedValue({
+        state: 'Completed',
+        result: { model: 'jev-1.13.0', answers, usage: { input_tokens: 426, output_tokens: 73 } },
+      }),
+    };
+
+    await instrumentWorkersAiClient(ai).run('typesafe/jev', { state: 'Help!', questions });
+
+    const span = spanToStaticSpanJSON(endedSpans[0]!);
+    expect(span.description).toBe('evaluate typesafe/jev');
+    expect(span.data).toEqual({
+      [SEMANTIC_ATTRIBUTE_SENTRY_ORIGIN]: 'auto.ai.cloudflare.workers_ai',
+      [SEMANTIC_ATTRIBUTE_SENTRY_OP]: 'gen_ai.evaluate',
+      [SEMANTIC_ATTRIBUTE_SENTRY_SAMPLE_RATE]: 1,
+      [SENTRY_SEGMENT_NAME_SOURCE]: 'custom',
+      [GEN_AI_PROVIDER_NAME]: 'cloudflare.workers_ai',
+      [GEN_AI_OPERATION_NAME]: 'evaluate',
+      [GEN_AI_REQUEST_MODEL]: 'typesafe/jev',
+      [GEN_AI_RESPONSE_MODEL]: 'jev-1.13.0',
+      [GEN_AI_USAGE_INPUT_TOKENS]: 426,
+      [GEN_AI_USAGE_OUTPUT_TOKENS]: 73,
+      [GEN_AI_USAGE_TOTAL_TOKENS]: 499,
+      [GEN_AI_INPUT_MESSAGES]: JSON.stringify([{ type: 'evaluation', state: 'Help!', questions }]),
+      [GEN_AI_OUTPUT_MESSAGES]: JSON.stringify([{ type: 'evaluation', answers }]),
+    });
+  });
+
   describe('span names', () => {
     function setupClient(traceLifecycle: 'static' | 'stream'): Span[] {
       const client = new TestClient(
diff --git a/packages/server-utils/test/ai/lib/utils/workers-ai-utils.test.ts b/packages/server-utils/test/ai/lib/utils/workers-ai-utils.test.ts
index 70878dc1a96e..8d9f9d15b257 100644
--- a/packages/server-utils/test/ai/lib/utils/workers-ai-utils.test.ts
+++ b/packages/server-utils/test/ai/lib/utils/workers-ai-utils.test.ts
@@ -14,6 +14,7 @@ import {
   GEN_AI_REQUEST_TOP_K,
   GEN_AI_REQUEST_TOP_P,
   GEN_AI_OUTPUT_MESSAGES,
+  GEN_AI_RESPONSE_MODEL,
   GEN_AI_RESPONSE_TEXT,
   GEN_AI_RESPONSE_TOOL_CALLS,
   GEN_AI_SYSTEM_INSTRUCTIONS,
@@ -59,6 +60,10 @@ describe('workers-ai utils', () => {
       expect(getOperationName({ text: 'embed me' })).toBe('embeddings');
     });
 
+    it('returns "evaluate" for TypeSafe state and questions inputs', () => {
+      expect(getOperationName({ state: 'Help!', questions: {} })).toBe('evaluate');
+    });
+
     it('prefers "chat" when both messages and text are present', () => {
       expect(getOperationName({ messages: [{ role: 'user', content: 'Hi' }], text: 'embed me' })).toBe('chat');
     });
@@ -185,6 +190,29 @@ describe('workers-ai utils', () => {
   });
 
   describe('addResponseAttributes', () => {
+    it.each([
+      { name: 'wrapped in { state, result }', wrap: (body: object) => ({ state: 'Completed', result: body }) },
+      { name: 'unwrapped', wrap: (body: object) => body },
+    ])('records evaluate responses $name', ({ wrap }) => {
+      const { span, attributes } = createMockSpan();
+      const answers = { is_urgent: { type: 'noul', noul: 0.95 } };
+
+      addResponseAttributes(
+        span,
+        wrap({ model: 'jev-1.13.0', answers, usage: { input_tokens: 366, output_tokens: 50 } }),
+        true,
+        'evaluate',
+      );
+
+      expect(attributes).toEqual({
+        [GEN_AI_RESPONSE_MODEL]: 'jev-1.13.0',
+        [GEN_AI_USAGE_INPUT_TOKENS]: 366,
+        [GEN_AI_USAGE_OUTPUT_TOKENS]: 50,
+        [GEN_AI_USAGE_TOTAL_TOKENS]: 416,
+        [GEN_AI_OUTPUT_MESSAGES]: JSON.stringify([{ type: 'evaluation', answers }]),
+      });
+    });
+
     it('sets token usage and computes the total from input and output tokens', () => {
       const { span, attributes } = createMockSpan();