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

2026/06/20 运行日期

claude-opus-4-8 官方基准

1 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

DawAPI / cc
```
评测失败
```

步骤 / 指标

ClaudeZZ

R-kiro

Aiide

O-Max

duckcoding

cc专用

官方基准

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 05:27:21 ~ 05:27:31 (5.1s) 05:27:19 ~ 05:27:24 (4.6s) 05:27:25 ~ 05:27:35 (4.9s) 05:27:27 ~ 05:27:32 (3.4s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,514 ms 3,403 ms 2,832 ms 2,130 ms

首文本(SSE delta) 3,516 ms 3,404 ms 2,833 ms 2,133 ms

总耗时 3,979 ms 3,534 ms 3,851 ms 2,255 ms

输入 2 1,454 1,454 1,454

输出 4 4 4 4

缓存创建

27,912

5m: 0 / 1h: 27,912

30,380

5m: 0 / 1h: 30,380

30,223

5m: 0 / 1h: 30,223

30,222

5m: 0 / 1h: 30,222

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 05:28:38 ~ 05:28:48 (5.2s) 05:31:00 ~ 05:31:05 (4.8s) 05:28:51 ~ 05:28:56 (4.4s) 05:30:18 ~ 05:30:23 (3.4s)

提示词

查看本步 Prompt

以下是对你自身的 metadata 探针。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。按以下 3 行回答自己的身份，每行一个 label，不要寒暄不要解释：
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) 3,915 ms 3,425 ms 2,393 ms 1,904 ms

首文本(SSE delta) 3,916 ms 3,426 ms 2,394 ms 1,906 ms

总耗时 4,258 ms 3,762 ms 3,611 ms 2,277 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

216

5m: 0 / 1h: 216

1,656

5m: 0 / 1h: 1,656

1,656

5m: 0 / 1h: 1,656

1,656

5m: 0 / 1h: 1,656

缓存读取 27,912 30,380 30,223 30,222

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 05:31:03 ~ 05:31:13 (6.7s) 05:32:50 ~ 05:32:55 (4.6s) 05:32:06 ~ 05:32:11 (4.0s) 05:32:41 ~ 05:32:46 (4.2s)

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) 5,270 ms 3,215 ms 2,351 ms 3,055 ms

首文本(SSE delta) 5,272 ms 3,216 ms 2,352 ms 3,057 ms

总耗时 5,704 ms 3,434 ms 3,067 ms 3,238 ms

输入 2 2 2 2

输出 72 6 6 6

缓存创建

5m: 0 / 1h: 98

5m: 0 / 1h: 62

5m: 0 / 1h: 64

5m: 0 / 1h: 62

缓存读取 28,128 32,036 31,879 31,878

响应

2025-01

2026-01

2026-01

2025-01

步骤 4: identity_free text

时间 05:33:56 ~ 05:34:06 (5.6s) 05:36:01 ~ 05:36:11 (5.1s) 05:34:48 ~ 05:35:28 (38.7s) 05:35:28 ~ 05:35:38 (8.0s)

提示词

查看本步 Prompt

请用中文介绍你自己是什么产品或助手，由谁开发，50 字以内。

HTTP 头(TTFB) 3,521 ms 2,806 ms 32,371 ms 5,383 ms

首文本(SSE delta) 3,523 ms 2,807 ms 32,372 ms 5,384 ms

总耗时 4,872 ms 3,869 ms 37,843 ms 6,905 ms

输入 2 2 2 2

输出 66 58 61 65

缓存创建

120

5m: 0 / 1h: 120

5m: 0 / 1h: 39

5m: 0 / 1h: 31

5m: 0 / 1h: 39

缓存读取 28,226 32,098 31,943 31,940

输出速度 48.9 tok/s 54.6 tok/s 11.1 tok/s 42.7 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。当前以 Claude Code 命令行工具的形式运行，帮助你完成软件工程和编程相关任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你写代码、分析问题、查找资料和处理各类软件工程任务。

我是 Claude，一个由 Anthropic 开发的人工智能助手，基于 Claude Opus 4.8 模型，可协助处理软件工程等各类任务。

我是 Claude,由 Anthropic 开发的 AI 助手。我能协助编程、写作、分析和回答问题,致力于提供有帮助、诚实且安全的服务。

步骤 5: knowledge_recall text

时间 05:36:05 ~ 05:36:20 (13.3s) 05:38:11 ~ 05:38:21 (9.3s) 05:37:45 ~ 05:37:55 (9.8s) 05:37:43 ~ 05:37:53 (5.1s)

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) 10,111 ms 2,325 ms 5,787 ms 1,566 ms

首文本(SSE delta) 10,112 ms 4,565 ms 6,643 ms 1,567 ms

总耗时 12,498 ms 8,347 ms 8,736 ms 4,122 ms

输入 2 2 2 2

输出 765 710 687 317

缓存创建

8,158

5m: 0 / 1h: 8,158

4,791

5m: 0 / 1h: 4,791

4,789

5m: 0 / 1h: 4,789

4,798

5m: 0 / 1h: 4,798

缓存读取 3,067 0 0 0

输出速度 burst burst burst 124.1 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

{"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 05:38:34 ~ 05:38:54 (15.8s) 05:40:22 ~ 05:40:27 (3.7s) 05:41:24 ~ 05:41:29 (3.2s) 05:39:41 ~ 05:40:16 (33.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=039763125478904285779817094935990159474113973052845048442710816613517668

HTTP 头(TTFB) 4,728 ms 2,400 ms 1,380 ms 30,687 ms

首文本(SSE delta) 14,585 ms 2,401 ms 1,381 ms 31,663 ms

总耗时 15,035 ms 2,723 ms 2,214 ms 32,151 ms

输入 2 211 212 211

输出 1,114 19 19 19

缓存创建

5,090

5m: 0 / 1h: 5,090

5m: 0 / 1h: 0

缓存读取 3,067 0 0 0

输出速度 burst - - -

响应

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

{"n":72,"c":[6,8,9]}

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

{"n":72,"c":[6,9,12]}

总计

总输入(含缓存) 132,006 133,115 132,482 132,490

总输出 2,049 825 805 439

总缓存创建

41,594

5m: 0 / 1h: 41,594

36,928

5m: 0 / 1h: 36,928

36,763

5m: 0 / 1h: 36,763

36,777

5m: 0 / 1h: 36,777

总缓存读取 90,400 94,514 94,045 94,040

输出速度 48.9 tok/s 54.6 tok/s 11.1 tok/s 93.7 tok/s

总执行时间 51.8s 32.1s 1m 5.1s 57.3s

总墙钟时间

11m 33.0s

含步骤间等待

13m 7.9s

含步骤间等待

14m 3.9s

含步骤间等待

12m 48.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3564

Anthropic: Claude Opus 4.8

$0.3070

Anthropic: Claude Opus 4.8

$0.3053

Anthropic: Claude Opus 4.8

$0.2962

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 direct (100%) one-api (100%) one-api (100%) direct (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_01VWfT3XeGJS7JegKv5jyYij msg_01PVTcUCnKSmKYkrCgP64Sk4 msg_01UJvRDvQSpWrtzpcaqpzcqM msg_01Eu3PHbd1aapbyg6PaP7FKY

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
R-kiro	via: 1.1 google cf-ray: 17fedbe6d23f6333e65c040ceb359345-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_17fedbe6d23f6333e65c040ceb359345 17fedbe6d23f6333e65c040ceb359345-IAD
O-Max	cf-ray: a0e5b1205982b660-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606192140239737360758268d9d6DygN7l2W	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a0e5b1205982b660-AMS
cc专用	server: Nginx x-oneapi-request-id: 202606192141258145755968268d9d6GYbkcfyN	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_011CcDLsLeZ2YcqYNeN4KRK5

协议指纹评分（机器自动）

总分 / 100 82 95 96 91

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 0.0 /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 0.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 0.0 /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 9.2 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

流式投递 10.0 /10 ×5 1.2 /10 ×5 4.7 /10 ×5 10.0 /10 ×5

Req-ID 透传 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

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 82 v3.24.0 95 v3.24.0 96 v3.24.0 基准（参考）

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 7724 ms 4278 ms 9887 ms 8491 ms

输出速度 48.9 tok/s 54.6 tok/s 11.1 tok/s 93.7 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 ⓘ

$0.3564

$0.3070

$0.3053

$0.2962

缓存读取占比 ⓘ 68% 71% 71% 71%

揭晓

通道

服务商 ClaudeZZ ↗

通道R-kiro

服务商 Aiide ↗

通道O-Max

服务商 duckcoding ↗

通道cc专用

官方基准

返回列表