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

2026/06/28 运行日期

claude-opus-4-8 官方基准

2 通道分数 < 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

官方基准

HongMaCC

u-cc

ClaudeZZ

R-kiro

SudoCode

M-Max

SAIAi

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 08:46:32 ~ 08:46:37 (3.3s) 08:46:26 ~ 08:46:31 (3.1s) 08:46:24 ~ 08:46:34 (5.0s) 08:46:28 ~ 08:46:38 (4.9s) 08:46:30 ~ 08:46:35 (3.7s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 1,935 ms 1,958 ms 3,813 ms 3,327 ms 2,480 ms

首文本(SSE delta) 1,936 ms 1,959 ms 3,815 ms 3,329 ms 2,482 ms

总耗时 2,067 ms 1,959 ms 4,070 ms 3,555 ms 2,596 ms

输入 1,454 0 2 1,454 1,454

输出 4 19 4 4 4

缓存创建

5m: 0 / 1h: 0

27,913

5m: 0 / 1h: 27,913

30,222

5m: 0 / 1h: 30,222

30,222

5m: 0 / 1h: 30,222

缓存读取 30,222 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 08:48:58 ~ 08:49:03 (3.3s) 08:47:51 ~ 08:47:56 (3.7s) 08:47:37 ~ 08:47:42 (4.0s) 08:47:56 ~ 08:48:01 (4.8s) 08:49:56 ~ 08:50:01 (3.4s)

提示词

查看本步 Prompt

输出 3 行 metadata 描述你自己（当前 API 模型）。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。禁止任何额外字符：
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) 1,974 ms 2,639 ms 2,543 ms 3,051 ms 2,064 ms

首文本(SSE delta) 1,975 ms 2,640 ms 2,544 ms 3,052 ms 2,066 ms

总耗时 2,375 ms 2,640 ms 3,114 ms 3,483 ms 2,446 ms

输入 2 0 2 2 2

输出 28 179 28 28 28

缓存创建

1,643

5m: 0 / 1h: 1,643

5m: 0 / 1h: 0

203

5m: 0 / 1h: 203

1,643

5m: 0 / 1h: 1,643

5m: 0 / 1h: 0

缓存读取 30,222 0 27,913 30,222 31,865

输出速度 - burst - - -

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 08:50:30 ~ 08:50:35 (3.6s) 08:50:59 ~ 08:51:04 (3.3s) 08:49:54 ~ 08:49:59 (4.3s) 08:51:49 ~ 08:51:54 (4.1s) 08:53:58 ~ 08:54:03 (3.1s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,350 ms 2,040 ms 3,052 ms 2,853 ms 1,931 ms

首文本(SSE delta) 2,351 ms 2,041 ms 3,053 ms 2,854 ms 1,933 ms

总耗时 2,581 ms 2,041 ms 3,455 ms 3,092 ms 2,144 ms

输入 2 0 2 2 2

输出 6 45 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 0

5m: 0 / 1h: 89

5m: 0 / 1h: 54

5m: 0 / 1h: 62

缓存读取 31,865 0 28,116 31,865 31,865

输出速度 - burst - - -

响应

2026-01

2026-01

2025-03

2026-01

2026-01

步骤 4: identity_free text

时间 08:53:08 ~ 08:53:13 (4.6s) 08:54:00 ~ 08:54:05 (3.8s) 08:51:16 ~ 08:51:26 (6.4s) 08:55:30 ~ 08:55:35 (4.7s) 08:57:44 ~ 08:57:49 (4.4s)

提示词

查看本步 Prompt

用中文说说你的身份：什么产品？由谁打造？50 字以内。

HTTP 头(TTFB) 2,899 ms 2,581 ms 4,748 ms 2,881 ms 2,645 ms

首文本(SSE delta) 2,901 ms 2,582 ms 4,750 ms 2,882 ms 2,647 ms

总耗时 3,536 ms 2,582 ms 5,476 ms 3,736 ms 3,437 ms

输入 2 0 2 2 2

输出 46 43 67 49 57

缓存创建

5m: 0 / 1h: 35

5m: 0 / 1h: 0

5m: 0 / 1h: 46

5m: 0 / 1h: 37

5m: 0 / 1h: 35

缓存读取 31,919 0 28,205 31,919 31,927

输出速度 72.4 tok/s burst 92.3 tok/s 57.4 tok/s 72.2 tok/s

响应

我是 Claude，由 Anthropic 公司打造的 AI 助手，当前模型为 Claude Opus 4.8。

我是Claude，Anthropic开发的AI助手。

我是 Claude，一个由 Anthropic 开发的 AI 助手。当前模型为 Claude Opus 4.8，可帮助你完成编程、写作、分析等各类任务。

我是 Claude，属于 Claude 系列产品，由 Anthropic 开发。当前模型为 Claude Opus 4.8。

我是 Claude，由 Anthropic 打造的 AI 助手，当前模型为 Claude Opus 4.8，可用于编程及多种任务协助。

步骤 5: knowledge_recall text

时间 08:56:30 ~ 08:56:40 (7.8s) 08:56:14 ~ 08:56:24 (8.4s) 08:55:01 ~ 08:55:11 (6.1s) 08:56:43 ~ 08:56:53 (9.3s) 09:01:30 ~ 09:01:40 (8.8s)

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) 1,212 ms 7,060 ms 2,917 ms 2,462 ms 1,962 ms

首文本(SSE delta) 4,525 ms 7,061 ms 2,918 ms 5,804 ms 5,275 ms

总耗时 6,900 ms 7,270 ms 5,203 ms 8,320 ms 7,755 ms

输入 2 0 2 2 2

输出 680 1,438 320 673 691

缓存创建

4,754

5m: 0 / 1h: 4,754

5m: 0 / 1h: 0

11,132

5m: 0 / 1h: 11,132

4,759

5m: 0 / 1h: 4,759

3,001

5m: 0 / 1h: 3,001

缓存读取 0 0 0 0 1,772

输出速度 burst burst 140.0 tok/s burst burst

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 09:00:35 ~ 09:00:40 (2.5s) 08:59:10 ~ 08:59:20 (5.7s) 08:57:25 ~ 08:57:41 (11.2s) 08:59:15 ~ 08:59:21 (3.3s) 09:04:56 ~ 09:05:01 (2.7s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=551613304837335754726393916680651795295653261081018369008741843216870421

HTTP 头(TTFB) 1,231 ms 4,431 ms 2,868 ms 1,973 ms 1,132 ms

首文本(SSE delta) 1,233 ms 4,432 ms 10,190 ms 1,974 ms 1,133 ms

总耗时 1,565 ms 4,581 ms 10,362 ms 2,387 ms 1,456 ms

输入 211 0 2 211 211

输出 19 1,031 879 19 19

缓存创建

5m: 0 / 1h: 0

5,091

5m: 0 / 1h: 5,091

5m: 0 / 1h: 0

缓存读取 0 0 3,067 0 0

输出速度 - burst burst - -

响应

{"n":72,"c":[8,9,7]}

{"n":69,"c":[6,6,5]}

{"n":72,"c":[7,6,5]}

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

{"n":72,"c":[8,7,6]}

总计

总输入(含缓存) 132,387 0 131,787 132,394 132,422

总输出 783 2,755 1,304 779 805

总缓存创建

6,486

5m: 0 / 1h: 6,486

5m: 0 / 1h: 0

44,474

5m: 0 / 1h: 44,474

36,715

5m: 0 / 1h: 36,715

33,320

5m: 0 / 1h: 33,320

总缓存读取 124,228 0 87,301 94,006 97,429

输出速度 72.4 tok/s burst 128.5 tok/s 57.4 tok/s 72.2 tok/s

总执行时间 25.1s 27.9s 36.9s 31.1s 26.2s

总墙钟时间

14m 8.0s

含步骤间等待

12m 54.0s

含步骤间等待

11m 16.9s

含步骤间等待

12m 52.9s

含步骤间等待

18m 31.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1306

Anthropic: Claude Opus 4.8

>=$0.0689

Anthropic: Claude Opus 4.8 (成本不完整)

$0.3543

Anthropic: Claude Opus 4.8

$0.3043

Anthropic: Claude Opus 4.8

$0.2855

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

ID 格式 msg_01TA69uAmWX4oNNv3Xo6Yksb msg_275fe36bd5d741c6ac85044a msg_01C88XsVL7dDwCuQKaJZ7Axp msg_01Kvbc5zE79go7YJ9EtzqwyY msg_01T7XxFxBTsJ4s5KNDdGDXGE

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
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_011CcUkfBA52CDQbnupn8TNC
u-cc	server: nginx x-new-api-version: 1.0.17 x-oneapi-request-id: 202606280059111674705498268d9d6nFb73bHI	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-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
R-kiro	via: 1.1 google cf-ray: 00e5a0aea44d6c8dd1b33960f4edcc63-IAD server: nginx	id_format: msg_prefix via_chain: ["1.1 google"] 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	req_00e5a0aea44d6c8dd1b33960f4edcc63 00e5a0aea44d6c8dd1b33960f4edcc63-IAD
M-Max	server: nginx x-new-api-version: 0b8be5cf x-oneapi-request-id: 20260628085917318254714ABZmX2j	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-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
O-Max	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_011CcUkzSx7sVTgWMrkB5x5S

协议指纹评分（机器自动）

总分 / 100 100 53 79 98 100

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 10.0 /10 ×20 0.0 /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 5.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 5.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 未计入 0.0 /10 ×8 10.0 /10 ×8 10.0 /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 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 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 9.7 /10 ×5 9.9 /10 ×5 10.0 /10 ×5

流式投递 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

与基线相似度基准（参考） 53 v3.25.0 79 v3.25.0 98 v3.25.0 100 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 3171 ms 3512 ms 5280 ms 4096 ms 3306 ms

输出速度 72.4 tok/s burst 128.5 tok/s 57.4 tok/s 72.2 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 ⓘ

$0.1306

>=$0.0689 (成本不完整)

$0.3543

$0.3043

$0.2855

缓存读取占比 ⓘ 94% - 66% 71% 74%

揭晓

通道

官方基准

服务商 HongMaCC ↗

通道u-cc

服务商 ClaudeZZ ↗

通道R-kiro

服务商 SudoCode ↗

通道M-Max

服务商 SAIAi ↗

通道O-Max

返回列表