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

2026/06/17 运行日期

claude-opus-4-8 官方基准

1 任务失败执行不完整

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

InfAI / R-Kiro
```
评测失败
```

步骤 / 指标

ClaudeCN

O-Max

RunAPI

O-Max

官方基准

duckcoding

cc专用

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 21:44:34 ~ 21:44:44 (6.3s) 21:44:34 ~ 21:44:44 (8.3s) 21:44:40 ~ 21:44:45 (3.9s) 21:44:36 ~ 21:44:51 (10.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,630 ms 5,944 ms 2,510 ms 8,384 ms

首文本(SSE delta) 3,635 ms 5,945 ms 2,512 ms 8,386 ms

总耗时 3,895 ms 6,387 ms 2,635 ms 8,755 ms

输入 1,454 1,454 1,454 1,454

输出 4 4 4 4

缓存创建

30,225

5m: 0 / 1h: 30,225

30,224

5m: 0 / 1h: 30,224

5m: 0 / 1h: 0

30,223

5m: 0 / 1h: 30,223

缓存读取 0 0 30,222 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 21:47:46 ~ 21:47:56 (6.7s) 21:48:33 ~ 21:48:38 (4.7s) 21:47:50 ~ 21:47:55 (3.8s) 21:47:02 ~ 21:47:17 (13.1s)

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) 4,125 ms 2,372 ms 2,394 ms 11,806 ms

首文本(SSE delta) 4,127 ms 3,320 ms 2,396 ms 11,807 ms

总耗时 4,902 ms 3,484 ms 2,812 ms 12,195 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

1,643

5m: 0 / 1h: 1,643

1,643

5m: 0 / 1h: 1,643

1,643

5m: 0 / 1h: 1,643

1,643

5m: 0 / 1h: 1,643

缓存读取 30,225 30,224 30,222 30,223

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 21:50:45 ~ 21:50:56 (5.0s) 21:50:30 ~ 21:50:40 (6.0s) 21:50:26 ~ 21:50:31 (3.6s) 21:50:11 ~ 21:50:21 (8.0s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,401 ms 3,553 ms 2,402 ms 6,359 ms

首文本(SSE delta) 3,402 ms 3,902 ms 2,404 ms 6,360 ms

总耗时 3,948 ms 3,976 ms 2,603 ms 6,802 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 64

5m: 0 / 1h: 62

缓存读取 31,868 31,867 31,865 31,866

响应

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 21:52:43 ~ 21:52:53 (6.2s) 21:53:16 ~ 21:53:26 (5.0s) 21:51:50 ~ 21:51:55 (3.8s) 21:54:21 ~ 21:54:31 (5.1s)

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) 3,352 ms 2,148 ms 1,988 ms 2,852 ms

首文本(SSE delta) 3,353 ms 2,150 ms 1,988 ms 2,853 ms

总耗时 5,108 ms 3,749 ms 2,785 ms 4,020 ms

输入 2 2 2 2

输出 58 63 62 63

缓存创建

5m: 0 / 1h: 35

5m: 0 / 1h: 39

5m: 0 / 1h: 31

缓存读取 31,922 31,921 31,929 31,928

输出速度 33.0 tok/s 39.4 tok/s 77.8 tok/s 54.0 tok/s

响应

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、写作与各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手，基于 Claude Opus 4.8 模型。我可以协助编程、写作、分析等多种任务。

我是 Claude，一个由 Anthropic 开发的人工智能助手。当前模型为 Claude Opus 4.8（API 标识 claude-opus-4-8）。

我是 Claude，一个由 Anthropic 开发的人工智能助手。我基于 Claude Opus 4.8 模型，API 标识为 claude-opus-4-8。

步骤 5: knowledge_recall text

时间 21:56:36 ~ 21:56:51 (10.6s) 21:54:35 ~ 21:54:50 (10.2s) 21:53:07 ~ 21:53:22 (8.3s) 21:55:59 ~ 21:56:10 (8.7s)

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) 2,700 ms 2,862 ms 1,820 ms 2,365 ms

首文本(SSE delta) 6,013 ms 5,798 ms 5,687 ms 5,213 ms

总耗时 8,371 ms 8,410 ms 7,408 ms 7,691 ms

输入 2 2 2 2

输出 686 696 705 677

缓存创建

4,769

5m: 0 / 1h: 4,769

4,777

5m: 0 / 1h: 4,777

3,004

5m: 0 / 1h: 3,004

4,776

5m: 0 / 1h: 4,776

缓存读取 0 0 1,772 0

输出速度 burst burst 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

步骤 6: digit_count text

时间 21:59:15 ~ 21:59:25 (5.2s) 21:57:51 ~ 21:58:01 (5.9s) 21:54:47 ~ 21:54:52 (3.9s) 21:58:25 ~ 21:58:35 (5.3s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=684121934925558794306991695655813587915083773339257771456467237053990613

HTTP 头(TTFB) 3,591 ms 3,094 ms 2,550 ms 3,932 ms

首文本(SSE delta) 3,592 ms 3,823 ms 2,551 ms 3,933 ms

总耗时 4,055 ms 4,009 ms 2,920 ms 4,326 ms

输入 211 211 211 212

输出 19 19 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

响应

{"n":72,"c":[2,7,12]}

{"n":72,"c":[2,8,12]}

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

{"n":72,"c":[4,8,12]}

总计

总输入(含缓存) 132,414 132,422 132,425 132,426

总输出 801 816 824 797

总缓存创建

36,726

5m: 0 / 1h: 36,726

36,737

5m: 0 / 1h: 36,737

4,742

5m: 0 / 1h: 4,742

36,735

5m: 0 / 1h: 36,735

总缓存读取 94,015 94,012 126,010 94,017

输出速度 33.0 tok/s 39.4 tok/s 77.8 tok/s 54.0 tok/s

总执行时间 40.1s 40.0s 27.3s 50.7s

总墙钟时间

14m 51.3s

含步骤间等待

13m 26.9s

含步骤间等待

10m 12.2s

含步骤间等待

13m 58.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3049

Anthropic: Claude Opus 4.8

$0.3054

Anthropic: Claude Opus 4.8

$0.1216

Anthropic: Claude Opus 4.8

$0.3049

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 - - -

ID 格式 msg_013szsCbvXCF1ukr4XLy7VEY msg_01HW5gB34XogTdV8bXVuCbU7 msg_01VUi9ozuDEXrxMrz73kV2ry msg_0143maEduMTBbVaGwwQ8qK82

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	cf-ray: a0d292e9ab3a1cae-AMS server: cloudflare x-oneapi-request-id: 20260617215917360220467UWXodbCR	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a0d292e9ab3a1cae-AMS
O-Max	server: nginx/1.24.0 (Ubuntu) x-oneapi-request-id: 2026061721575341080291sfD8eHCm	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_011Cc8woN36Mn2yBxhvUAr2A
cc专用	server: Nginx x-oneapi-request-id: 202606171358264429772108268d9d60FiodOlO	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	—

协议指纹评分（机器自动）

总分 / 100 98 98 100 97

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 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 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

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 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

系统提示纯净 10.0 /10 ×8 10.0 /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

知识截止 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

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 10.0 /10 ×5 10.0 /10 ×5 9.2 /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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度 98 v3.22.0 98 v3.22.0 基准（参考） 97 v3.22.0

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 自由身份 identity_free_clean
6 停止原因 stop_reason_present
7 系统提示纯净 system_prompt_clean
8 Service Tier service_tier_present
9 缓存命中比 cache_hit_ratio_match
10 缓存 TTL 一致 cache_ttl_consistency
11 推理地区 inference_geo_present
12 缓存连续 cache_continuity_intra
13 知识召回 knowledge_recall_match
14 延迟基线 latency_baseline_match
15 原生 msg-ID anthropic_msg_id_format
16 Sliding 缓存 cache_sliding_correctness
17 结构化身份 identity_structured_match
18 信封自报一致 envelope_self_report_match
19 中文遵循 instruction_following_lang
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 5047 ms 5003 ms 3527 ms 7298 ms

输出速度 33.0 tok/s 39.4 tok/s 77.8 tok/s 54.0 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 ⓘ

$0.3049

$0.3054

$0.1216

$0.3049

缓存读取占比 ⓘ 71% 71% 95% 71%

揭晓

通道

服务商 ClaudeCN ↗

通道O-Max

服务商 RunAPI ↗

通道O-Max

官方基准

服务商 duckcoding ↗

通道cc专用

返回列表