模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

前 5 步同一会话，第 6 步独立新会话，覆盖 6 个独立信号： 1. ping/pong —— 单字指令遵循 + 建立缓存上下文。 2. 身份（结构化）—— 三行 vendor / brand / model 格式，机器可解析。 3. 知识截止 —— 跨越 5 分钟 cache 边界后追问，检测 sliding 5m cache 是否真正命中。 4. 身份（自由格式）—— 自然语言自报身份，给包装层（Kiro 等）暴露品牌的机会。 5. 世界知识层级 —— 5 道公共事件硬事实题，按答对档位映射实测 tier，与请求模型对照。 6. 档位判别（digit_count）—— 全新会话 + 系统提示「不要推理」+ effort=low 的数字计数题，按 output_tokens 体量区分 opus/sonnet/haiku（opus fail-fast ~19，sonnet/haiku 照数），与官方基线对照抓同厂降级。前 5 步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；总跨度 > 6 分钟。第 6 步是独立新会话（不复用前 5 轮上下文，否则 opus 不再 fail-fast）。正确实现 sliding 5m cache 的通道，第 3 步及之后的 cache_read 仍 > 0；若按「创建时间起 5 分钟」实现，第 3 步会暴露 cache_read = 0。

3 通道 · 6 步骤

2026/06/25 运行日期

claude-opus-4-8 官方基准

3 任务失败执行不完整

已揭晓揭晓进度

⚠ 2 通道不可测 · 已从主对比表移除点击展开原因

YunWu / O-Api
```
评测失败
```
Aiberm / cc
```
评测失败
```

步骤 / 指标

PackyCode

cc-Zeus

官方基准

SSSAiCode

O-Max

基本信息

服务名 Claude Code Claude Code Claude Code

模型 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8

客户端 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 23:46:21 ~ 23:46:31 (6.1s) 23:46:25 ~ 23:46:30 (3.4s) 23:46:23 ~ 23:46:33 (7.2s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,663 ms 1,913 ms 5,471 ms

首文本(SSE delta) 3,843 ms 1,915 ms 5,600 ms

总耗时 3,851 ms 2,065 ms 6,001 ms

输入 1,454 1,454 1,428

输出 4 4 4

缓存创建

30,225

5m: 0 / 1h: 30,225

30,222

5m: 0 / 1h: 30,222

29,758

5m: 0 / 1h: 29,758

缓存读取 0 0 0

响应

pong

pong

pong

步骤 2: identity text

时间 23:48:44 ~ 23:48:54 (5.1s) 23:49:59 ~ 23:50:04 (3.8s) 23:48:26 ~ 23:48:31 (3.9s)

提示词

查看本步 Prompt

本请求询问的是你（正在生成此回复的当前 API 模型）自身的身份，不是用户提到的任何外部产品。必须直接作答，禁止追问、拒答、解释或输出占位符；不能要求更多上下文。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
vendor: <厂商公司名一个英文小写单词>
brand: <产品或模型品牌名一个英文小写单词>
model: <实际 API model 字段值，保留原始短横线格式>

HTTP 头(TTFB) 3,652 ms 2,240 ms 2,337 ms

首文本(SSE delta) 3,695 ms 2,241 ms 2,702 ms

总耗时 4,154 ms 2,757 ms 2,901 ms

输入 2 2 2

输出 28 28 30

缓存创建

1,646

5m: 0 / 1h: 1,646

1,646

5m: 0 / 1h: 1,646

1,620

5m: 0 / 1h: 1,620

缓存读取 30,225 30,222 29,758

响应

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-20250514

步骤 3: cutoff text

时间 23:51:54 ~ 23:52:04 (5.8s) 23:53:43 ~ 23:53:48 (3.9s) 23:51:52 ~ 23:51:57 (4.0s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,767 ms 2,226 ms 2,726 ms

首文本(SSE delta) 2,768 ms 2,227 ms 2,916 ms

总耗时 3,195 ms 2,942 ms 3,182 ms

输入 2 2 2

输出 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 56

缓存读取 31,871 31,868 31,378

响应

2026-01

2026-01

2025-03

步骤 4: identity_free text

时间 23:55:20 ~ 23:55:30 (5.6s) 23:57:25 ~ 23:57:30 (3.8s) 23:54:46 ~ 23:54:56 (5.2s)

提示词

查看本步 Prompt

用中文回答：你是谁？由谁开发？50 字以内。

HTTP 头(TTFB) 2,979 ms 2,056 ms 2,679 ms

首文本(SSE delta) 2,980 ms 2,060 ms 3,552 ms

总耗时 4,540 ms 2,814 ms 3,710 ms

输入 2 2 2

输出 59 47 70

缓存创建

5m: 0 / 1h: 31

5m: 0 / 1h: 37

5m: 0 / 1h: 39

缓存读取 31,925 31,922 31,434

输出速度 37.8 tok/s 62.3 tok/s -

响应

我是 Claude，由 Anthropic 开发的人工智能助手，具体模型是 Claude Opus 4.8，可协助处理软件工程等各类任务。

我是 Claude，由 Anthropic 开发，属于 Claude 模型系列，当前为 Opus 4.8 版本。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮助你完成编程、写作、分析等各类任务，致力于提供有帮助、诚实且无害的回应。

步骤 5: knowledge_recall text

时间 23:59:29 ~ 23:59:44 (10.1s) 23:59:39 ~ 23:59:49 (5.3s) 23:56:44 ~ 23:56:54 (7.0s)

提示词

查看本步 Prompt

Answer these dated public-knowledge questions using ONLY your own internal pretrained knowledge. Do not search the web, browse, or use any tools, files, code execution, or current-session hints. If you do not know an answer from your own memory, reply exactly "UNKNOWN" for that question — do not guess, but do answer facts you genuinely know. Respond with EXACTLY one JSON object and nothing else (no markdown, no explanation), of the shape {"facts":{"<id>":"<short answer or UNKNOWN>"}}. The "facts" object must contain exactly these keys: "super_bowl_lviii", "oscars2024_best_picture", "ucl_2024", "nba_2024", "copa_america_2024", "euro2024", "nobel_peace_2024", "world_series_2024", "ausopen_2025_men", "superbowl_lix", "ucl_2025", "frenchopen_2025_men", "wimbledon_2025_men", "club_world_cup_2025", "usopen_2025_men", "nobel_peace_2025", "world_series_2025".
- "super_bowl_lviii": Which NFL team won Super Bowl LVIII (played 2024-02-11)?
- "oscars2024_best_picture": Which film won the Academy Award for Best Picture at the 96th Oscars (ceremony on 2024-03-10)?
- "ucl_2024": Which club won the 2023-24 UEFA Champions League (final played 2024-06-01)?
- "nba_2024": Which team won the 2024 NBA Finals (concluded 2024-06-17)?
- "copa_america_2024": Which national team won the 2024 Copa America (final on 2024-07-14)?
- "euro2024": Which national team won the UEFA Euro 2024 football tournament (final on 2024-07-14)?
- "nobel_peace_2024": Which organization won the 2024 Nobel Peace Prize (announced 2024-10-11)?
- "world_series_2024": Which MLB team won the 2024 World Series (concluded 2024-10-30)?
- "ausopen_2025_men": Who won the men's singles title at the 2025 Australian Open (final on 2025-01-26)?
- "superbowl_lix": Which NFL team won Super Bowl LIX (played 2025-02-09)?
- "ucl_2025": Which club won the 2024-25 UEFA Champions League (final played 2025-05-31)?
- "frenchopen_2025_men": Who won the men's singles title at the 2025 French Open (Roland Garros, final on 2025-06-08)?
- "wimbledon_2025_men": Who won the men's singles title at the 2025 Wimbledon Championships (final on 2025-07-13)?
- "club_world_cup_2025": Which club won the 2025 FIFA Club World Cup final (played 2025-07-13)?
- "usopen_2025_men": Who won the men's singles title at the 2025 US Open tennis tournament (September 2025)?
- "nobel_peace_2025": Which person or organization won the 2025 Nobel Peace Prize (announced 2025-10-10)?
- "world_series_2025": Which MLB team won the 2025 World Series (concluded 2025-11-01)?

HTTP 头(TTFB) 2,706 ms 1,490 ms 2,823 ms

首文本(SSE delta) 6,935 ms 1,493 ms 3,358 ms

总耗时 8,246 ms 4,336 ms 6,055 ms

输入 2 2 2

输出 695 329 340

缓存创建

4,769

5m: 0 / 1h: 4,769

4,760

5m: 0 / 1h: 4,760

4,297

5m: 0 / 1h: 4,297

缓存读取 0 0 0

输出速度 burst 115.7 tok/s 126.1 tok/s

响应

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel

步骤 6: digit_count text

时间 06/26 00:00:50 ~ 00:00:55 (3.4s) 06/26 00:02:58 ~ 00:03:03 (3.2s) 06/26 00:00:29 ~ 00:00:34 (1.4s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=476359783669607890919337069611296137428161744160885250958086263957595453

HTTP 头(TTFB) 2,073 ms 1,769 ms -

首文本(SSE delta) 2,265 ms 1,772 ms -

总耗时 2,466 ms 2,210 ms -

输入 212 211 -

输出 19 19 -

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 -

响应

{"n":72,"c":[3,8,11]}

{"n":72,"c":[3,8,10]}

失败

评测失败

总计

总输入(含缓存)

132,420

132,404

129,776

仅前 6 步

总输出

811

433

450

仅前 6 步

总缓存创建

36,725

5m: 0 / 1h: 36,725

36,719

5m: 0 / 1h: 36,719

35,770

5m: 0 / 1h: 35,770

仅前 6 步

总缓存读取

94,021

94,012

92,570

仅前 6 步

输出速度

37.8 tok/s

104.5 tok/s

126.1 tok/s

仅前 6 步

总执行时间

36.1s

23.5s

28.7s

仅前 6 步

总墙钟时间

14m 34.2s

含步骤间等待

16m 38.2s

含步骤间等待

14m 11.0s

含步骤间等待

仅前 6 步

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3052

Anthropic: Claude Opus 4.8

$0.2957

Anthropic: Claude Opus 4.8

$0.2883

Anthropic: Claude Opus 4.8 (部分)

通道指纹

来源步骤 digit_count digit_count knowledge_recall

平台 one-api (100%) direct (100%) direct (100%)

上游 Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN cloudflare - -

ID 格式 msg_01Q4tHwCxbj3rDhRZ5xodTvA msg_01Tuz7nDfyym3XW8B2hieTPP msg_01HLDAaQkE8ukdDhnd9QCXsh

展开详细指纹（响应头 / body 特征 / 完整信号列表）

通道

response_headers_notable

response_body_traits

signals

request_id_chain

cc-Zeus

via: 1.1 Caddy
cf-ray: a1153002eac1f7df-LAX
server: cloudflare
x-oneapi-request-id: 20260626000051681827412v9Vw7pNg

id_format: msg_prefix
via_chain: ["1.1 Caddy"]
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

hdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-rayhdr:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

a1153002eac1f7df-LAX

官方基准

server: nginx/1.22.1

id_format: msg_prefix
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

req_011CcQG3bRJCSAU6UqoDoKmx

O-Max

cf-ray: a1152a023cb42b92-LAX
server: Photon-Edge
set-cookie: _cfuvid=E4MqH28rX6WDytqhH_7TNvHr2ZANRV9oLH5wiGpT6X0-1782403005.8022246-1.0.1.1-u30Y_vk7vctMEOWbqYDTry4j0dWWiahSjRfWhnVUEOw; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com

id_format: msg_prefix
extra_fields: ["stop_details"]
has_service_tier: true
has_inference_geo: true
model_has_date_suffix: false
usage_has_cache_fields: true

body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body

1782403005388-ded2mj29y8eb8
req_011CcQFa2YcU6wyF1uroim4G
a1152a023cb42b92-LAX

协议指纹评分（机器自动）

总分 / 100 97 98

基线状态已对照基线已对照基线

缓存命中比 10.0 /10 ×20 10.0 /10 ×20 —

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 —

模型匹配 10.0 /10 ×14 10.0 /10 ×14 —

缓存连续性 10.0 /10 ×14 10.0 /10 ×14 —

sliding 5m 10.0 /10 ×13 10.0 /10 ×13 —

原生 msg-ID 10.0 /10 ×8 10.0 /10 ×8 —

系统提示纯净 10.0 /10 ×8 10.0 /10 ×8 —

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 —

知识截止 10.0 /10 ×7 10.0 /10 ×7 —

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 —

service_tier 10.0 /10 ×6 10.0 /10 ×6 —

inference_geo 10.0 /10 ×5 10.0 /10 ×5 —

延迟基线 10.0 /10 ×5 10.0 /10 ×5 —

流式投递 8.6 /10 ×5 10.0 /10 ×5 —

Req-ID 透传 0.0 /10 ×4 10.0 /10 ×4 —

stop_reason 10.0 /10 ×3 10.0 /10 ×3 —

SDK 一致 10.0 /10 ×2 10.0 /10 ×2 —

综合结论

与基线相似度 97 v3.24.1 基准（参考）执行失败，不评分

雷达图例

共 22 个维度（顺时针，从顶部 12 点起）— 点击展开对照

1 模型匹配 model_match
2 知识截止 cutoff_match
3 SDK 一致 sdk_consistency
4 Thinking 存在 thinking_present
5 流式投递 buffer_dump_match
6 自由身份 identity_free_clean
7 停止原因 stop_reason_present
8 系统提示纯净 system_prompt_clean
9 Service Tier service_tier_present
10 缓存命中比 cache_hit_ratio_match
11 缓存 TTL 一致 cache_ttl_consistency
12 推理地区 inference_geo_present
13 缓存连续 cache_continuity_intra
14 知识召回 knowledge_recall_match
15 延迟基线 latency_baseline_match
16 原生 msg-ID anthropic_msg_id_format
17 Sliding 缓存 cache_sliding_correctness
18 结构化身份 identity_structured_match
19 信封自报一致 envelope_self_report_match
20 中文遵循 instruction_following_lang
21 档位思考量 tier_thinking_volume_match
22 Req-ID 透传 anthropic_request_id_passthrough

每根轴长度 = 该维度 0-10 分（越长越接近基线）。蓝色虚线圆 = 满分基准。具体权重见方法论页。

维度雷达 —

平均延迟 4409 ms 2854 ms 4370 ms

输出速度 37.8 tok/s 104.5 tok/s 126.1 tok/s

完成步骤 6/6 成功 6/6 成功 5/6 成功 · 1 失败

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3052

$0.2957

$0.2883 (部分)

缓存读取占比 ⓘ 71% 71% 71% (部分)

揭晓

通道

服务商 PackyCode ↗

通道cc-Zeus

官方基准

服务商 SSSAiCode ↗

通道O-Max

返回列表