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。

5 通道 · 6 步骤

2026/06/19 运行日期

claude-opus-4-8 官方基准

全部 ≥ 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

RunAPI

O-Max

FoxCode

官方基准

PackyCode

cc-Zeus

0-0

O-Max

基本信息

服务名 Claude Code 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 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 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 21:58:31 ~ 21:58:36 (4.8s) 21:58:27 ~ 21:58:32 (4.4s) 21:58:33 ~ 21:58:38 (3.5s) 21:58:25 ~ 21:58:35 (5.1s) 21:58:29 ~ 21:58:39 (7.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,207 ms 3,024 ms 2,011 ms 2,052 ms 4,611 ms

首文本(SSE delta) 2,209 ms 3,026 ms 2,013 ms 2,054 ms 4,857 ms

总耗时 2,456 ms 3,277 ms 2,226 ms 3,002 ms 4,929 ms

输入 1,454 1,454 1,454 1,454 2

输出 4 4 4 4 4

缓存创建

30,224

5m: 0 / 1h: 30,224

30,222

5m: 30,222 / 1h: 0

30,222

5m: 0 / 1h: 30,222

30,223

5m: 0 / 1h: 30,223

32,326

5m: 0 / 1h: 32,326

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 22:01:16 ~ 22:01:21 (4.7s) 22:00:03 ~ 22:00:08 (4.9s) 22:00:50 ~ 22:00:55 (3.4s) 21:59:36 ~ 21:59:41 (4.5s) 22:00:06 ~ 22:00:16 (8.9s)

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。请只输出以下 3 行内容（label 小写，值小写）：
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) 2,266 ms 3,558 ms 2,025 ms 1,709 ms 6,652 ms

首文本(SSE delta) 2,274 ms 3,559 ms 2,027 ms 1,710 ms 6,728 ms

总耗时 2,744 ms 4,079 ms 2,400 ms 2,238 ms 6,728 ms

输入 2 2 2 2 2

输出 28 28 28 28 28

缓存创建

31,853

5m: 0 / 1h: 31,853

1,630

5m: 1,630 / 1h: 0

1,630

5m: 0 / 1h: 1,630

1,630

5m: 0 / 1h: 1,630

3,718

5m: 0 / 1h: 3,718

缓存读取 0 30,222 30,222 30,223 28,785

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

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

步骤 3: cutoff text

时间 22:04:03 ~ 22:04:08 (4.9s) 22:02:10 ~ 22:02:15 (3.6s) 22:03:02 ~ 22:03:07 (3.2s) 22:02:32 ~ 22:02:37 (3.5s) 22:02:44 ~ 22:02:54 (6.5s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,376 ms 2,378 ms 1,821 ms 2,043 ms 3,928 ms

首文本(SSE delta) 3,377 ms 2,396 ms 1,823 ms 2,276 ms 3,930 ms

总耗时 3,561 ms 2,718 ms 2,042 ms 2,466 ms 5,300 ms

输入 2 2 2 2 2

输出 6 6 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 54 / 1h: 0

5m: 0 / 1h: 54

3,782

5m: 0 / 1h: 3,782

缓存读取 31,853 31,852 31,852 31,853 28,785

响应

2026-01

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 22:08:08 ~ 22:08:18 (5.3s) 22:05:20 ~ 22:05:25 (4.3s) 22:05:43 ~ 22:05:48 (3.6s) 22:05:05 ~ 22:05:10 (3.5s) 22:06:28 ~ 22:06:38 (7.6s)

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) 2,082 ms 2,119 ms 1,748 ms 1,666 ms 4,275 ms

首文本(SSE delta) 2,083 ms 2,120 ms 1,749 ms 1,668 ms 4,277 ms

总耗时 3,363 ms 3,321 ms 2,648 ms 2,249 ms 5,404 ms

输入 2 2 2 2 2

输出 72 64 59 42 69

缓存创建

5m: 0 / 1h: 39

5m: 31 / 1h: 0

5m: 0 / 1h: 35

5m: 0 / 1h: 37

3,816

5m: 0 / 1h: 3,816

缓存读取 31,907 31,906 31,906 31,907 28,785

输出速度 56.3 tok/s 53.3 tok/s 65.6 tok/s 72.3 tok/s 61.2 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你写代码、解答问题、分析文档和处理各类软件工程任务。当前模型为 Opus 4.8。

我是 Claude，由 Anthropic 开发的 AI 助手。当前模型为 Claude Opus 4.8，可协助完成软件工程、信息查询等多种任务。

我是 Claude，由 Anthropic 打造的 AI 助手，当前模型为 Claude Opus 4.8，可协助编程、研究、写作等任务。

我是 Claude（claude-opus-4-8），属于 Claude 产品系列，由 Anthropic 开发。

我是 Claude，由 Anthropic 打造的 AI 助手。当前模型为 Claude Opus 4.8，可作为 Claude Code 在终端等环境协助软件工程任务。

步骤 5: knowledge_recall text

时间 22:11:17 ~ 22:11:32 (10.9s) 22:09:09 ~ 22:09:19 (9.9s) 22:08:18 ~ 22:08:28 (8.4s) 22:07:46 ~ 22:08:01 (10.2s) 22:08:38 ~ 22:08:53 (13.2s)

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) 2,659 ms 3,042 ms 2,208 ms 2,537 ms 4,780 ms

首文本(SSE delta) 6,461 ms 6,867 ms 5,035 ms 6,410 ms 7,048 ms

总耗时 8,692 ms 9,070 ms 7,388 ms 7,873 ms 11,593 ms

输入 2 2 2 2 2

输出 656 719 655 667 674

缓存创建

4,772

5m: 0 / 1h: 4,772

4,755

5m: 4,755 / 1h: 0

4,754

5m: 0 / 1h: 4,754

4,740

5m: 0 / 1h: 4,740

5,354

5m: 0 / 1h: 5,354

缓存读取 0 0 0 0 0

输出速度 burst burst burst burst 148.3 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

{"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 22:13:10 ~ 22:13:15 (2.9s) 22:10:50 ~ 22:10:55 (3.1s) 22:12:23 ~ 22:12:28 (3.2s) 22:11:05 ~ 22:11:10 (4.3s) 22:10:39 ~ 22:10:49 (5.3s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=631328524774594351254844204555101205700462753659013751997642333971685185

HTTP 头(TTFB) 1,162 ms 1,987 ms 1,967 ms 1,393 ms 2,711 ms

首文本(SSE delta) 1,163 ms 1,988 ms 1,968 ms 1,394 ms 2,712 ms

总耗时 1,834 ms 2,348 ms 2,280 ms 1,979 ms 3,066 ms

输入 211 211 211 212 794

输出 19 19 19 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0 0

响应

{"n":72,"c":[7,5,3]}

{"n":72,"c":[6,5,4]}

{"n":72,"c":[6,4,5]}

{"n":72,"c":[5,5,4]}

{"n":72,"c":[5,5,5]}

总计

总输入(含缓存) 132,375 132,345 132,348 132,341 136,155

总输出 785 840 771 766 800

总缓存创建

66,942

5m: 0 / 1h: 66,942

36,692

5m: 36,692 / 1h: 0

36,695

5m: 0 / 1h: 36,695

36,684

5m: 0 / 1h: 36,684

48,996

5m: 0 / 1h: 48,996

总缓存读取 63,760 93,980 93,980 93,983 86,355

输出速度 56.3 tok/s 53.3 tok/s 65.6 tok/s 72.3 tok/s 131.0 tok/s

总执行时间 33.6s 30.1s 25.4s 31.0s 48.5s

总墙钟时间

14m 43.9s

含步骤间等待

12m 28.0s

含步骤间等待

13m 54.9s

含步骤间等待

12m 45.0s

含步骤间等待

12m 20.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4783

Anthropic: Claude Opus 4.8

$0.2197

Anthropic: Claude Opus 4.8

$0.3040

Anthropic: Claude Opus 4.8

$0.3038

Anthropic: Claude Opus 4.8

$0.3734

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count digit_count

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

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

CDN - cloudflare - cloudflare -

ID 格式 msg_01GKkMvGnLuc2Xv2BwHBFCMT msg_01WFchmA7cVkYFRwxfgQ6xSD msg_01BauNTUEBmSwuYZ48VnrtYn msg_014BEYDJczrZjQmzwrm2CCeF msg_01TRYi29eh4fX3Y2NyQ6Exgj

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
O-Max	server: nginx/1.24.0 (Ubuntu) x-oneapi-request-id: 20260619221311403521118LJXkk7jL	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	—
cc	cf-ray: a0e31e9cdf6ee041-NRT server: cloudflare set-cookie: _cfuvid=5c4cVq4RKiA8N48NiDrcJkRI917Wrt83uIifdAIp1gM-1781878251.3703368-1.0.1.1-20m7GGQFdiPCvFPddkukovgGLo6wgtn8._zYlWTTeuY; 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	hdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011CcCke7YSjTDRd6cmcuiUv a0e31e9cdf6ee041-NRT
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_011CcCkkyYBVBvzru4Geiz1r
cc-Zeus	via: 1.1 Caddy cf-ray: a0e31f0508202283-LAX server: cloudflare x-oneapi-request-id: 202606192211077124904312vlry4Ol	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	a0e31f0508202283-LAX
O-Max	via: 1.1 Caddy server: uvicorn, nginx/1.24.0 (Ubuntu)	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:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a69938e9ec4b686a4d41a5d658d8840e

协议指纹评分（机器自动）

总分 / 100 91 92 100 96 95

基线状态已对照基线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 6.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

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

模型匹配 10.0 /10 ×14 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 10.0 /10 ×14

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

原生 msg-ID 10.0 /10 ×8 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 10.0 /10 ×8 5.0 /10 ×8

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

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

service_tier 10.0 /10 ×6 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 10.0 /10 ×5

延迟基线 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 9.4 /10 ×5

流式投递 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 6.0 /10 ×5 10.0 /10 ×5

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

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

SDK 一致 10.0 /10 ×2 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 91 v3.24.0 92 v3.24.0 基准（参考） 96 v3.24.0 95 v3.24.0

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 3775 ms 4136 ms 3164 ms 3301 ms 6170 ms

输出速度 56.3 tok/s 53.3 tok/s 65.6 tok/s 72.3 tok/s 131.0 tok/s

完成步骤 6/6 成功 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 ⓘ

$0.4783

$0.2197

$0.3040

$0.3038

$0.3734

缓存读取占比 ⓘ 48% 71% 71% 71% 63%

揭晓

通道

服务商 RunAPI ↗

通道O-Max

服务商 FoxCode ↗

通道cc

官方基准

服务商 PackyCode ↗

通道cc-Zeus

服务商 0-0 ↗

通道O-Max

返回列表