diff --git a/content/blog/2026-10-15-llm-as-judge/data-process.svg b/content/blog/2026-10-15-llm-as-judge/data-process.svg
new file mode 100644
index 0000000000000..9d9a860511ed6
--- /dev/null
+++ b/content/blog/2026-10-15-llm-as-judge/data-process.svg
@@ -0,0 +1,64 @@
+
diff --git a/content/blog/2026-10-15-llm-as-judge/evaluation-record-list.png b/content/blog/2026-10-15-llm-as-judge/evaluation-record-list.png
new file mode 100644
index 0000000000000..3691461978903
Binary files /dev/null and b/content/blog/2026-10-15-llm-as-judge/evaluation-record-list.png differ
diff --git a/content/blog/2026-10-15-llm-as-judge/index.md b/content/blog/2026-10-15-llm-as-judge/index.md
new file mode 100644
index 0000000000000..04dfd1e542724
--- /dev/null
+++ b/content/blog/2026-10-15-llm-as-judge/index.md
@@ -0,0 +1,207 @@
+---
+title: "How Good Are LLM Answers? Bringing Model Quality into SkyWalking Observability"
+author: "peachisai"
+date: 2026-10-15
+description: "Learn how the upcoming SkyWalking 11.1.0 release uses LLM-as-Judge to evaluate GenAI span input and output messages asynchronously and connect the results to evaluation records, metrics, and traces."
+tags:
+ - AI
+ - Metrics
+ - Tracing
+---
+
+## Introduction
+
+Traditional observability focuses on service error rates and latency. These signals do not tell us whether an LLM response is grounded in its context, relevant to the question, or complete. A knowledge-base assistant may return a well-formatted and confident answer without evidence. The trace can still be successful while users consider the answer unreliable. In an AI gateway or agent, model versions, provider health, and routing policies can also change and affect response quality.
+
+SkyWalking already provides OpenTelemetry-based GenAI observability for request volume, success rate, latency, tokens, and estimated cost. The upcoming SkyWalking 11.1.0 release adds LLM-as-Judge on top of this foundation. For sampled GenAI spans, it sends the input and output messages to an OpenAI-compatible judge model, produces configurable evaluation results, and associates them with an evaluation record and the source trace. SCORE tasks also produce a metric for the Virtual GenAI Dashboard and alarms.
+
+## Why use LLM-as-Judge with SkyWalking?
+
+- Datasets and benchmarks are useful for model selection, prompt comparison, and regression testing before a release. They usually use fixed questions and contexts, however, and cannot continuously represent production response quality.
+- Production requests keep changing. User questions and contexts differ, model versions are updated, and provider status and traffic patterns fluctuate. Production systems therefore need continuous evaluation of real requests in addition to offline tests.
+- LLM-as-Judge generates quality results for real requests, while SkyWalking places those results in the existing observability path. A trace shows the LLM interactions and external calls involved in a request, helping engineers determine whether a quality regression relates to an earlier LLM output, an MCP call, or another external response.
+- The Virtual GenAI Dashboard and alarm capabilities show quality trends over time and can alert on abnormal values. This connects quality evaluation, trend monitoring, alerting, and trace investigation into a production observability loop.
+
+## Data flow
+
+
+
+Evaluation is asynchronous: eligible GenAI spans are queued, and trace ingestion never waits for the judge. The module only processes GenAI spans that contain both non-empty `gen_ai.input.messages` and `gen_ai.output.messages`; spans missing either attribute are skipped.
+
+## Usage
+
+This guide targets SkyWalking 11.1.0. Until that version is released, use an `11.1.0-SNAPSHOT` development build of OAP. The evaluation pages shown below are in the `main` branch of Horizon UI and not in Horizon UI 1.0.1, so use the `ghcr.io/apache/skywalking-horizon-ui:main` image until Horizon UI 1.1.0 is released.
+
+### 1. Enable the module
+
+The module is disabled by default. Set `SW_AI_EVALUATION=default` when starting OAP to enable it. The module settings in `application.yml` are:
+
+```yaml
+ai-evaluation:
+ selector: ${SW_AI_EVALUATION:-}
+ default:
+ sampleRate: ${SW_AI_EVALUATION_SAMPLE_RATE:1000000}
+ bufferSize: ${SW_AI_EVALUATION_BUFFER_SIZE:100}
+ consumerThreads: ${SW_AI_EVALUATION_CONSUMER_THREADS:8}
+ # Maximum escaped characters included from each input/output message field.
+ maxContentLength: ${SW_AI_EVALUATION_MAX_CONTENT_LENGTH:16384}
+```
+
+`sampleRate` is expressed in parts per million (PPM): `1000000` evaluates every eligible trace and `10000` evaluates approximately 1%. Sampling uses the trace ID, so spans from the same trace are selected consistently. When the queue is full, new evaluation tasks are dropped instead of blocking trace ingestion. Tune `bufferSize` and `consumerThreads` only after checking judge capacity and OAP resource usage.
+
+Setting the selector alone is not sufficient. OAP also needs a valid endpoint, model, API key, and system prompt in `ai-evaluation.yml`. The evaluator remains inactive when these required settings are missing.
+
+### 2. Configure the judge and tasks
+
+Configure the judge in `ai-evaluation.yml`:
+
+```yaml
+judge:
+ provider: openai
+ endpoint: ${AI_EVALUATION_ENDPOINT:}
+ model: ${AI_EVALUATION_MODEL:gpt-5.5}
+ api-key: ${AI_EVALUATION_API_KEY:}
+ request-timeout-seconds: 30
+ max-retries: 2
+ temperature: 0.2
+ max_tokens: 100000
+```
+
+The `provider` currently supports only `openai`. The endpoint may be any service that accepts OpenAI Chat Completions requests and returns the judge's JSON result in `choices[0].message.content`. Judge requests contain the input and output messages from the span. Before enabling this in production, review data redaction, privacy requirements, the provider's data handling policy, and model usage cost.
+
+The system prompt must require a JSON response. The default prompt is:
+
+```yaml
+system-prompt: |
+ You are an AI observability evaluator.
+
+ Your task is to evaluate GenAI spans according to the evaluation tasks provided by the user.
+
+ General Rules:
+
+ 1. Return ONLY valid JSON.
+ 2. Do NOT return markdown.
+ 3. Do NOT return code fences.
+ 4. Do NOT return explanations outside the JSON result.
+ 5. Do NOT add any fields that are not requested.
+ 6. Do NOT omit any evaluation task.
+ 7. The JSON keys MUST appear in exactly the same order as the evaluation tasks provided by the user.
+ 8. Each evaluation task MUST contain:
+ - value
+ - reason
+ 9. The reason field should be concise and explain the evaluation result in one or two sentences.
+ 10. If the available context is insufficient, provide the best evaluation possible and explain the limitation in the reason field.
+
+ Value Type Rules:
+
+ SCORE
+ - value must be a number between 0.0 and 1.0.
+ - 1.0 represents the best possible evaluation result.
+ - 0.0 represents the worst possible evaluation result.
+
+ BOOLEAN
+ - value must be either true or false.
+
+ STRING
+ - value must be a string.
+
+ JSON
+ - value must be a valid JSON object.
+
+ Evaluation Principles:
+
+ - Evaluate only using the information provided in the span context.
+ - Do not assume facts that are not present in the input.
+ - Do not invent missing information.
+ - Be objective and consistent.
+ - Use the task instruction as the primary evaluation criterion.
+
+ Return ONLY the JSON object.
+```
+
+Use `level` to define levels for SCORE and BOOLEAN results. A SCORE range includes `min` and excludes `max`, except that a range with `max: 1.0` also includes `1.0`:
+
+```yaml
+level:
+ undefined: undefined
+ score:
+ - min: 0.0
+ max: 0.3
+ level: fail
+ - min: 0.3
+ max: 0.6
+ level: warning
+ - min: 0.6
+ max: 0.85
+ level: good
+ - min: 0.85
+ max: 1.0
+ level: excellent
+ boolean:
+ "true": excellent
+ "false": fail
+```
+
+Each task requires `name`, `valueType`, and `instruction`. Supported value types are `SCORE`, `BOOLEAN`, `STRING`, and `JSON`. Evaluation records keep every valid task result, and invalid results are skipped. SCORE and BOOLEAN values are stored as numbers in PPM (a score of `0.8` becomes `800000`, `true` becomes `1000000`, and `false` becomes `0`), and STRING and JSON values are stored as strings. Only SCORE tasks additionally report the `gen_ai_model_evaluation_score_ppm` metric; BOOLEAN, STRING, and JSON tasks are not included in this metric.
+
+```yaml
+tasks:
+ - name: Faithfulness
+ valueType: SCORE
+ instruction: Evaluate factual grounding. Higher is better.
+
+ - name: Relevance
+ valueType: SCORE
+ instruction: Evaluate response relevance. Higher is better.
+
+ - name: TaskCompletion
+ valueType: BOOLEAN
+ instruction: Evaluate task completion.
+
+ - name: Hallucination
+ valueType: SCORE
+ instruction: Evaluate factuality and absence of hallucinations. Higher is better.
+```
+
+The system prompt treats `1.0` as the best result, and the levels above map higher scores to better levels, so write every SCORE instruction so that a higher score means a better answer. That is why the Hallucination task above scores the absence of hallucinations rather than their severity.
+
+Every task result contains `value` and `reason`. SCORE values in both the records and the metric are in PPM, so divide by `1000000` when displaying the original score. You can define custom evaluation dimensions and use `allowedValues` for STRING tasks.
+
+### 3. Receive GenAI traces
+
+Start OAP and send GenAI traces to SkyWalking. See the [GenAI observability guide](../2026-04-05-virtual-genai-monitoring/) for instrumentation details. Evaluation needs the message content, and the SkyWalking Java agent does not collect it by default. For Spring AI instrumented by the Java agent, enable it explicitly in the environment of the application:
+
+```shell
+export SW_PLUGIN_SPRINGAI_COLLECT_INPUT_MESSAGES=true
+export SW_PLUGIN_SPRINGAI_COLLECT_OUTPUT_MESSAGES=true
+```
+
+For OTLP and Zipkin ingestion, make sure the instrumentation exports `gen_ai.input.messages` and `gen_ai.output.messages`. A span missing either attribute does not enter the evaluation queue.
+
+For OTLP ingestion, OAP 11.1.0 stores spans natively by default (`otlpTraceStorage: otlp`) and evaluates those spans directly, without converting them into Zipkin form.
+
+### 4. View evaluation records
+
+
+
+In Horizon UI, open **Virtual GenAI > Evaluation records** and filter records by task, value type, trace ID, and other fields. Records also retain the service, provider, model, operation, task, value, level, reason, judge model, and evaluation time.
+
+### 5. Follow an evaluation to its trace
+
+Click a record to view its details:
+
+
+
+For a record from a SkyWalking native trace, like the one below, users with the `traces:read` permission can click the **trace** link in the upper-right corner to open the source trace and inspect the recorded agent interactions, LLM calls, and external tool calls.
+
+
+
+Horizon UI `main` does not yet support following an OTLP evaluation record to its natively stored source trace. This UI limitation does not prevent evaluation records from being generated or queried.
+
+### 6. View quality metrics in the Dashboard
+
+
+
+The Models dashboard of Virtual GenAI shows the `gen_ai_model_evaluation_score_ppm` metric. In Virtual GenAI, the provider is the service and the model is its instance, so the chart shows SCORE trends per model and task. You can also create alarms from the metric.
+
+Also monitor the OAP self-observability metrics `ai_evaluation_dropped_count` and `ai_evaluation_error_count`, stored as `meter_oap_instance_ai_evaluation_dropped_count` and `meter_oap_instance_ai_evaluation_error_count`, so evaluation problems are not mistaken for model quality regressions. Their `reason` label tells the causes apart: for dropped evaluations, `pipeline_capacity` means the queue was full and `incomplete_span` means the evaluation context or its trace ID was missing, or a span lacked the required messages or its service, provider, model, or trace reference information; errors are split into `rejected`, `timeout`, and `invalid_response`.
diff --git a/content/blog/2026-10-15-llm-as-judge/jump-to-trace.png b/content/blog/2026-10-15-llm-as-judge/jump-to-trace.png
new file mode 100644
index 0000000000000..66f81944b53d9
Binary files /dev/null and b/content/blog/2026-10-15-llm-as-judge/jump-to-trace.png differ
diff --git a/content/blog/2026-10-15-llm-as-judge/model-dashboard.png b/content/blog/2026-10-15-llm-as-judge/model-dashboard.png
new file mode 100644
index 0000000000000..31a809f9d3be2
Binary files /dev/null and b/content/blog/2026-10-15-llm-as-judge/model-dashboard.png differ
diff --git a/content/blog/2026-10-15-llm-as-judge/record-detail.png b/content/blog/2026-10-15-llm-as-judge/record-detail.png
new file mode 100644
index 0000000000000..4077ba58bf88e
Binary files /dev/null and b/content/blog/2026-10-15-llm-as-judge/record-detail.png differ
diff --git a/content/zh/2026-10-15-llm-as-judge/data-process.svg b/content/zh/2026-10-15-llm-as-judge/data-process.svg
new file mode 100644
index 0000000000000..80fbdfda11fec
--- /dev/null
+++ b/content/zh/2026-10-15-llm-as-judge/data-process.svg
@@ -0,0 +1,64 @@
+
diff --git a/content/zh/2026-10-15-llm-as-judge/evaluation-record-list.png b/content/zh/2026-10-15-llm-as-judge/evaluation-record-list.png
new file mode 100644
index 0000000000000..3691461978903
Binary files /dev/null and b/content/zh/2026-10-15-llm-as-judge/evaluation-record-list.png differ
diff --git a/content/zh/2026-10-15-llm-as-judge/index.md b/content/zh/2026-10-15-llm-as-judge/index.md
new file mode 100644
index 0000000000000..7e7f74bf03730
--- /dev/null
+++ b/content/zh/2026-10-15-llm-as-judge/index.md
@@ -0,0 +1,207 @@
+---
+title: "LLM 回答得好不好？SkyWalking 用 LLM-as-Judge 评估线上回答"
+author: "peachisai"
+date: 2026-10-15
+description: "SkyWalking 11.1.0 即将支持 LLM-as-Judge：异步评估 GenAI Span 中的输入和输出消息，生成评估记录并关联原始 Trace，SCORE 任务还会产生质量指标。"
+tags:
+ - AI
+ - Metrics
+ - Tracing
+---
+
+## 介绍
+
+传统可观测性关注服务的错误率和延迟，但这些指标说明不了 LLM 的回答有没有上下文依据、是否切题、是否完整。知识库助手可能给出格式规范、语气笃定却没有依据的答案：Trace 显示请求成功，用户却觉得答案靠不住。在 AI 网关和探针中，模型版本、模型服务商的运行状态和路由策略也会变化，进而影响回答质量。
+
+SkyWalking 已经基于 OpenTelemetry 提供 GenAI 可观测性，可以查看调用量、成功率、延迟、Token 用量和预估费用。即将发布的 SkyWalking 11.1.0 在此基础上加入了 LLM-as-Judge。OAP 从采样到的 GenAI Span 中取出输入和输出消息，发给一个 OpenAI-compatible 的 Judge 模型。Judge 按配置的任务评估回答质量，OAP 把结果存为评估记录，并关联到原始 Trace。`SCORE` 类型的任务还会生成指标，供 Virtual GenAI Dashboard 和告警使用。
+
+## 为什么把质量评估接入 SkyWalking
+
+- 数据集和 Benchmark 适合用在模型选型、Prompt 对比和发布前的回归测试上。不过它们的问题和上下文通常是固定的，没法持续反映线上的回答质量。
+- 线上请求一直在变：用户问题和上下文各不相同，模型版本会更新，模型服务商的状态和流量模式也在波动。所以生产系统除了离线测试，还需要持续评估真实请求。
+- LLM-as-Judge 负责评估真实请求的回答质量，SkyWalking 则把评估结果和已有的 Trace、指标放在一起。Trace 能展示一次请求经过了哪些 LLM 交互和外部调用；回答质量下降时，可以据此判断问题是否和前序 LLM 的输出、MCP 调用或其他外部接口的返回有关。
+- Virtual GenAI Dashboard 和告警可以按时间展示质量趋势，并在数值异常时告警。这样，质量评估、趋势监控、告警和 Trace 排查就连成了一个线上观测闭环。
+
+## 数据流转
+
+
+
+评估是异步的：符合条件的 GenAI Span 先进入队列，Trace 接收不用等待 Judge 返回。只有 `gen_ai.input.messages` 和 `gen_ai.output.messages` 都非空的 GenAI Span 才会参与评估，缺任何一个就直接跳过。
+
+## 接入步骤
+
+本文以 SkyWalking 11.1.0 为准。正式发布前，OAP 请使用 `11.1.0-SNAPSHOT` 开发版本。下文截图中的评估页面目前只在 Horizon UI 的 `main` 分支中，Horizon UI 1.0.1 还没有；在 Horizon UI 1.1.0 发布前，请使用 `ghcr.io/apache/skywalking-horizon-ui:main` 镜像。
+
+### 1. 开启模块
+
+模块默认关闭，启动 OAP 时设置 `SW_AI_EVALUATION=default` 即可开启。`application.yml` 中的模块配置如下：
+
+```yaml
+ai-evaluation:
+ selector: ${SW_AI_EVALUATION:-}
+ default:
+ sampleRate: ${SW_AI_EVALUATION_SAMPLE_RATE:1000000}
+ bufferSize: ${SW_AI_EVALUATION_BUFFER_SIZE:100}
+ consumerThreads: ${SW_AI_EVALUATION_CONSUMER_THREADS:8}
+ # Maximum escaped characters included from each input/output message field.
+ maxContentLength: ${SW_AI_EVALUATION_MAX_CONTENT_LENGTH:16384}
+```
+
+`sampleRate` 的单位是百万分之一（PPM）：`1000000` 表示评估所有符合条件的 Trace，`10000` 约为 1%。采样按 Trace ID 进行，同一条 Trace 中的 Span 采样结果一致。队列满了以后，OAP 会直接丢弃新的评估任务，不会阻塞 Trace 接收。调整 `bufferSize` 和 `consumerThreads` 之前，先确认 Judge 服务能承受多少请求，再看看 OAP 的资源占用。
+
+只设置 `selector` 还不够，还要在 `ai-evaluation.yml` 中配置有效的 `endpoint`、模型、API Key 和系统 Prompt。缺少其中任何一项，评估模块都不会工作。
+
+### 2. 配置 Judge 和评估任务
+
+在 `ai-evaluation.yml` 中配置 Judge：
+
+```yaml
+judge:
+ provider: openai
+ endpoint: ${AI_EVALUATION_ENDPOINT:}
+ model: ${AI_EVALUATION_MODEL:gpt-5.5}
+ api-key: ${AI_EVALUATION_API_KEY:}
+ request-timeout-seconds: 30
+ max-retries: 2
+ temperature: 0.2
+ max_tokens: 100000
+```
+
+目前 `provider` 只支持 `openai`。`endpoint` 可以指向任何兼容 OpenAI Chat Completions 的服务：它要能接受 Chat Completions 请求，并在响应的 `choices[0].message.content` 中返回 Judge 的 JSON 结果。Judge 请求会带上 Span 中的输入和输出消息。在生产环境启用前，请先确认数据脱敏和隐私要求，了解服务提供方的数据处理政策，并估算模型调用费用。
+
+系统 Prompt 必须要求 Judge 返回 JSON。默认配置如下：
+
+```yaml
+system-prompt: |
+ You are an AI observability evaluator.
+
+ Your task is to evaluate GenAI spans according to the evaluation tasks provided by the user.
+
+ General Rules:
+
+ 1. Return ONLY valid JSON.
+ 2. Do NOT return markdown.
+ 3. Do NOT return code fences.
+ 4. Do NOT return explanations outside the JSON result.
+ 5. Do NOT add any fields that are not requested.
+ 6. Do NOT omit any evaluation task.
+ 7. The JSON keys MUST appear in exactly the same order as the evaluation tasks provided by the user.
+ 8. Each evaluation task MUST contain:
+ - value
+ - reason
+ 9. The reason field should be concise and explain the evaluation result in one or two sentences.
+ 10. If the available context is insufficient, provide the best evaluation possible and explain the limitation in the reason field.
+
+ Value Type Rules:
+
+ SCORE
+ - value must be a number between 0.0 and 1.0.
+ - 1.0 represents the best possible evaluation result.
+ - 0.0 represents the worst possible evaluation result.
+
+ BOOLEAN
+ - value must be either true or false.
+
+ STRING
+ - value must be a string.
+
+ JSON
+ - value must be a valid JSON object.
+
+ Evaluation Principles:
+
+ - Evaluate only using the information provided in the span context.
+ - Do not assume facts that are not present in the input.
+ - Do not invent missing information.
+ - Be objective and consistent.
+ - Use the task instruction as the primary evaluation criterion.
+
+ Return ONLY the JSON object.
+```
+
+可以在 `level` 中为 `SCORE` 和 `BOOLEAN` 结果配置等级。`SCORE` 区间包含 `min`、不包含 `max`；`max` 为 `1.0` 的区间也包含 `1.0` 本身：
+
+```yaml
+level:
+ undefined: undefined
+ score:
+ - min: 0.0
+ max: 0.3
+ level: fail
+ - min: 0.3
+ max: 0.6
+ level: warning
+ - min: 0.6
+ max: 0.85
+ level: good
+ - min: 0.85
+ max: 1.0
+ level: excellent
+ boolean:
+ "true": excellent
+ "false": fail
+```
+
+每个评估任务都要设置 `name`、`valueType` 和 `instruction`。`valueType` 支持 `SCORE`、`BOOLEAN`、`STRING` 和 `JSON`。评估记录保存所有合法的任务结果，不合法的结果直接跳过。`SCORE` 和 `BOOLEAN` 按 PPM 存为数值（分数 `0.8` 存为 `800000`，`true` 存为 `1000000`，`false` 存为 `0`），`STRING` 和 `JSON` 按字符串保存。只有 `SCORE` 任务会额外上报 `gen_ai_model_evaluation_score_ppm` 指标，其他类型只生成评估记录。
+
+```yaml
+tasks:
+ - name: Faithfulness
+ valueType: SCORE
+ instruction: Evaluate factual grounding. Higher is better.
+
+ - name: Relevance
+ valueType: SCORE
+ instruction: Evaluate response relevance. Higher is better.
+
+ - name: TaskCompletion
+ valueType: BOOLEAN
+ instruction: Evaluate task completion.
+
+ - name: Hallucination
+ valueType: SCORE
+ instruction: Evaluate factuality and absence of hallucinations. Higher is better.
+```
+
+系统 Prompt 把 `1.0` 当作最好的结果，上面的等级规则也是分数越高、等级越好，所以每个 `SCORE` 任务的 `instruction` 都要写成“分数越高，回答越好”。这也是为什么上面的 Hallucination 任务评的是“有没有幻觉”，而不是“幻觉有多严重”。
+
+每个任务的结果都包含 `value` 和 `reason`。评估记录和指标中的 `SCORE` 值都是 PPM，展示原始分数时除以 `1000000` 即可。你也可以自定义评估维度，并为 `STRING` 任务配置 `allowedValues`。
+
+### 3. 接收 GenAI Trace
+
+启动 OAP，并向 SkyWalking 上报 GenAI Trace，接入方法见 [GenAI 可观测性接入指南](../2026-04-05-virtual-genai-monitoring/)。评估需要消息内容，而 SkyWalking Java 探针默认不采集。用 Java 探针接入 Spring AI 时，需要在业务应用的启动环境中显式开启：
+
+```shell
+export SW_PLUGIN_SPRINGAI_COLLECT_INPUT_MESSAGES=true
+export SW_PLUGIN_SPRINGAI_COLLECT_OUTPUT_MESSAGES=true
+```
+
+通过 OTLP 或 Zipkin 接入时，请确认采集端会导出 `gen_ai.input.messages` 和 `gen_ai.output.messages`。缺少任一属性的 Span 不会进入评估队列。
+
+通过 OTLP 接入时，OAP 11.1.0 默认原生存储 Span（`otlpTraceStorage: otlp`），并直接评估这些 Span，无需转换为 Zipkin 格式。
+
+### 4. 查看评估记录
+
+
+
+在 Horizon UI 中打开 **Virtual GenAI > Evaluation records**，可以按任务、值类型、Trace ID 等条件筛选记录。每条记录还保留了服务、模型服务商、模型、操作、任务、评估值、等级、原因、Judge 模型和评估时间。
+
+### 5. 从评估记录查看原始 Trace
+
+点击一条记录，查看评估详情：
+
+
+
+记录来自 SkyWalking 原生 Trace 时，有 `traces:read` 权限即可点击右上角的 **trace** 链接，打开原始 Trace（如下图）。在 Trace 中可以查看采集到的 Agent 交互、LLM 调用和外部工具调用。
+
+
+
+Horizon UI 的 `main` 分支目前还不支持从 OTLP 评估记录跳转到原生存储的原始 Trace。这项 UI 限制不影响评估记录的生成和查询。
+
+### 6. 在 Dashboard 中查看质量指标
+
+
+
+Virtual GenAI 的 Models 页面展示了 `gen_ai_model_evaluation_score_ppm` 指标。在 Virtual GenAI 中，模型服务商对应服务，模型对应服务实例，所以这里看到的是按模型和任务划分的 `SCORE` 趋势。你也可以基于这个指标配置告警。
+
+另外，还要持续关注 OAP 自观测指标 `ai_evaluation_dropped_count` 和 `ai_evaluation_error_count`（存储名分别为 `meter_oap_instance_ai_evaluation_dropped_count` 和 `meter_oap_instance_ai_evaluation_error_count`），以免把评估本身的问题误判为模型回答质量下降。具体原因看 `reason` 标签：丢弃原因有两种，`pipeline_capacity` 表示队列已满，`incomplete_span` 表示缺少评估上下文或 Trace ID，或者 Span 缺少所需的消息、服务、模型服务商、模型或 Trace 关联信息；错误原因分为 `rejected`、`timeout` 和 `invalid_response`。
diff --git a/content/zh/2026-10-15-llm-as-judge/jump-to-trace.png b/content/zh/2026-10-15-llm-as-judge/jump-to-trace.png
new file mode 100644
index 0000000000000..66f81944b53d9
Binary files /dev/null and b/content/zh/2026-10-15-llm-as-judge/jump-to-trace.png differ
diff --git a/content/zh/2026-10-15-llm-as-judge/model-dashboard.png b/content/zh/2026-10-15-llm-as-judge/model-dashboard.png
new file mode 100644
index 0000000000000..31a809f9d3be2
Binary files /dev/null and b/content/zh/2026-10-15-llm-as-judge/model-dashboard.png differ
diff --git a/content/zh/2026-10-15-llm-as-judge/record-detail.png b/content/zh/2026-10-15-llm-as-judge/record-detail.png
new file mode 100644
index 0000000000000..4077ba58bf88e
Binary files /dev/null and b/content/zh/2026-10-15-llm-as-judge/record-detail.png differ