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。

4 通道 · 6 步骤

2026/06/30 运行日期

claude-opus-4-8 官方基准

2 通道分数 < 70 另含 1 任务失败

已揭晓揭晓进度

⚠ 1 通道不可测 · 已从主对比表移除点击展开原因

TopRouterCN / O-Max
```
评测失败
```

步骤 / 指标

PackyCode

cc-Zeus

duckcoding

cc专用

官方基准

FoxCode

R-ultra

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-opus-4-8 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 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 08:17:51 ~ 08:17:56 (3.6s) 08:17:53 ~ 08:17:58 (4.6s) 08:17:59 ~ 08:18:09 (5.1s) 08:17:57 ~ 08:18:02 (3.2s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,441 ms 3,260 ms 3,837 ms 2,071 ms

首文本(SSE delta) 2,442 ms 3,266 ms 3,839 ms 2,197 ms

总耗时 2,675 ms 3,528 ms 3,932 ms 2,319 ms

输入 1,454 1,454 1,454 73

输出 4 4 4 2

缓存创建

30,224

5m: 0 / 1h: 30,224

30,223

5m: 0 / 1h: 30,223

30,222

5m: 0 / 1h: 30,222

23,487

5m: 23,487 / 1h: 0

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 08:18:58 ~ 08:19:03 (3.2s) 08:20:22 ~ 08:20:32 (5.1s) 08:19:46 ~ 08:19:51 (3.4s) 08:19:24 ~ 08:19:34 (5.6s)

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) 1,906 ms 3,500 ms 2,018 ms 3,431 ms

首文本(SSE delta) 1,908 ms 3,503 ms 2,020 ms 3,432 ms

总耗时 2,231 ms 4,114 ms 2,410 ms 4,685 ms

输入 2 2 2 28

输出 28 28 28 85

缓存创建

1,646

5m: 0 / 1h: 1,646

1,646

5m: 0 / 1h: 1,646

1,646

5m: 0 / 1h: 1,646

251

5m: 251 / 1h: 0

缓存读取 30,224 30,223 30,222 23,487

输出速度 - - - 67.8 tok/s

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-4-8

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

步骤 3: cutoff text

时间 08:22:16 ~ 08:22:26 (6.9s) 08:24:26 ~ 08:24:36 (7.6s) 08:20:52 ~ 08:20:57 (3.7s) 08:22:58 ~ 08:23:03 (3.5s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 2,251 ms 6,718 ms 2,327 ms 1,580 ms

首文本(SSE delta) 2,252 ms 6,720 ms 2,328 ms 1,581 ms

总耗时 5,826 ms 6,774 ms 2,562 ms 2,422 ms

输入 2 2 2 74

输出 6 6 6 54

缓存创建

5m: 0 / 1h: 62

31,931

5m: 0 / 1h: 31,931

5m: 0 / 1h: 64

24,122

5m: 24,122 / 1h: 0

缓存读取 31,870 0 31,868 0

输出速度 - - - 64.2 tok/s

响应

2026-01

2025-01

2026-01

2026-01

步骤 4: identity_free text

时间 08:24:21 ~ 08:24:26 (4.4s) 08:26:46 ~ 08:26:56 (8.4s) 08:22:08 ~ 08:22:13 (4.5s) 08:26:53 ~ 08:26:58 (4.7s)

提示词

查看本步 Prompt

中文做个自我介绍：你是什么产品或助手？谁开发的？50 字内。

HTTP 头(TTFB) 2,003 ms 6,093 ms 2,316 ms 2,422 ms

首文本(SSE delta) 2,004 ms 6,102 ms 2,317 ms 2,423 ms

总耗时 3,283 ms 7,380 ms 3,468 ms 3,749 ms

输入 2 2 2 21

输出 67 60 69 104

缓存创建

5m: 0 / 1h: 39

24,580

5m: 24,580 / 1h: 0

缓存读取 31,932 31,931 31,932 0

输出速度 52.4 tok/s 46.9 tok/s 59.9 tok/s 78.4 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我能协助编程、写作、分析和解答各类问题，致力于提供有帮助、诚实且安全的回应。

我是 Claude，由 Anthropic 开发的人工智能助手。我可以帮助你编写代码、解答问题、分析内容、处理各类软件工程任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮助处理软件工程、写作、分析等多种任务，致力于提供有益、诚实且安全的协助。

我是 Kiro，一款由 Anthropic 开发的 AI 编程助手，基于 Claude Opus 4.8 模型驱动，专注于帮助开发者完成软件工程、代码分析、规划等专业任务。

步骤 5: knowledge_recall text

时间 08:27:05 ~ 08:27:15 (8.7s) 08:29:41 ~ 08:29:56 (11.1s) 08:24:28 ~ 08:24:38 (8.6s) 08:27:58 ~ 08:28:28 (25.3s)

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) 1,476 ms 7,430 ms 1,504 ms 1,900 ms

首文本(SSE delta) 5,296 ms 7,434 ms 4,799 ms 1,901 ms

总耗时 7,722 ms 10,328 ms 7,581 ms 24,442 ms

输入 2 2 2 67

输出 690 320 688 1,996

缓存创建

4,792

5m: 0 / 1h: 4,792

4,784

5m: 0 / 1h: 4,784

4,794

5m: 0 / 1h: 4,794

4,790

5m: 4,790 / 1h: 0

缓存读取 0 0 0 0

输出速度 burst 110.6 tok/s burst 88.5 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

{"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 08:30:12 ~ 08:30:17 (3.2s) 08:32:43 ~ 08:32:48 (3.4s) 08:27:45 ~ 08:27:50 (3.3s) 08:30:05 ~ 08:30:25 (18.6s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=151456511172837131693043319736378859533365474445725963296467290110346683

HTTP 头(TTFB) 1,640 ms 1,884 ms 1,888 ms 1,899 ms

首文本(SSE delta) 1,641 ms 1,887 ms 1,890 ms 1,900 ms

总耗时 2,054 ms 2,379 ms 2,335 ms 17,890 ms

输入 212 212 211 19

输出 19 19 19 2,014

缓存创建

5m: 0 / 1h: 0

191

5m: 191 / 1h: 0

缓存读取 0 0 0 0

输出速度 - - - 126.0 tok/s

响应

{"n":70,"c":[3,8,5]}

{"n":70,"c":[3,6,4]}

{"n":70,"c":[3,8,5]}

```json
{"n":72,"c":[3,7,6]}
```

总计

总输入(含缓存) 132,463 132,451 132,460 101,190

总输出 814 437 814 4,255

总缓存创建

36,763

5m: 0 / 1h: 36,763

68,623

5m: 0 / 1h: 68,623

36,765

5m: 0 / 1h: 36,765

77,421

5m: 77,421 / 1h: 0

总缓存读取 94,026 62,154 94,022 23,487

输出速度 52.4 tok/s 91.1 tok/s 59.9 tok/s 101.4 tok/s

总执行时间 30.0s 40.2s 28.5s 1m 0.9s

总墙钟时间

12m 25.9s

含步骤间等待

14m 55.0s

含步骤间等待

9m 50.6s

含步骤间等待

12m 28.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3055

Anthropic: Claude Opus 4.8

$0.4793

Anthropic: Claude Opus 4.8

$0.3055

Anthropic: Claude Opus 4.8

$0.4220

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN cloudflare - - cloudflare

ID 格式 msg_01WY9jXMm3J7EmfurYhp4UYj msg_01GnGYFxDfoM6ZVpURtVPKJd msg_0184zrpeTuXfHpuWhohx6e4Y msg_01YGqBJmOk27PxeRTIHf8lEq

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: a1390fa89aee23de-LAX server: cloudflare x-oneapi-request-id: 20260630083013943481204W5PDCeyn	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	a1390fa89aee23de-LAX
cc专用	server: Nginx x-oneapi-request-id: 202606300032450384276228268d9d6XfR66oYA	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	hdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
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_011CcYVmomR898MGPkAxJiJS
R-ultra	cf-ray: a1390f79ee36e393-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606300030065797043108268d9d6EePDHXo6	id_format: msg_prefix has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsupstream:anthropic-direct-from-body	a1390f79ee36e393-NRT

协议指纹评分（机器自动）

总分 / 100 96 74 98 55

基线状态已对照基线已对照基线已对照基线已对照基线

缓存命中比 8.6 /10 ×20 0.0 /10 ×20 8.6 /10 ×20 0.0 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 0.0 /10 ×15

模型匹配 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

缓存连续性 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 7.7 /10 ×13 10.0 /10 ×13 0.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

知识截止 10.0 /10 ×7 0.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 0.0 /10 ×7

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

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

延迟基线 10.0 /10 ×5 6.3 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

流式投递 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 7.4 /10 ×5

Req-ID 透传 0.0 /10 ×4 0.0 /10 ×4 10.0 /10 ×4 0.0 /10 ×4

stop_reason 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3

SDK 一致 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2

综合结论

与基线相似度 96 v3.25.0 74 v3.25.0 基准（参考） 55 v3.25.0

雷达图例

共 21 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 档位思考量 tier_thinking_volume_match
21 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 3965 ms 5751 ms 3715 ms 9251 ms

输出速度 52.4 tok/s 91.1 tok/s 59.9 tok/s 101.4 tok/s

完成步骤 6/6 成功 6/6 成功 6/6 成功 6/6 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3055

$0.4793

$0.3055

$0.4220

缓存读取占比 ⓘ 71% 47% 71% 23%

揭晓

通道

服务商 PackyCode ↗

通道cc-Zeus

服务商 duckcoding ↗

通道cc专用

官方基准

服务商 FoxCode ↗

通道R-ultra

返回列表