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

2026/07/01 运行日期

claude-haiku-4-5-20251001 官方基准

3 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

SSSAiCode

R-Kiro

NowCoding

R-Kiro

ByteCatCode

O-Max

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code Claude Code

模型 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001

客户端 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195

偏差标红基于官方基准

步骤 1: ping text

时间 15:37:01 ~ 15:37:06 (3.6s) 15:36:59 ~ 15:37:09 (5.3s) 15:36:53 ~ 15:37:03 (8.1s) 15:36:55 ~ 15:37:05 (5.3s) 15:36:57 ~ 15:37:02 (4.2s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 1,786 ms 2,893 ms 1,741 ms 3,306 ms 2,338 ms

首文本(SSE delta) 1,788 ms 2,895 ms 1,743 ms 3,328 ms -

总耗时 2,228 ms 2,895 ms 3,731 ms 3,953 ms 2,808 ms

输入 10 0 0 21,419 10

输出 50 22 279 93 67

缓存创建

6,166

5m: 0 / 1h: 6,166

5m: 0 / 1h: 0

21,545

5m: 0 / 1h: 21,545

缓存读取 15,379 0 0 0 0

输出速度 113.6 tok/s - 140.3 tok/s 148.8 tok/s 142.6 tok/s

响应

pong

pong

Ready. No prior task to resume here—the conversation started with ping/pong. What should I work on?

pong

pong

步骤 2: identity text

时间 15:39:36 ~ 15:39:46 (5.0s) 15:38:57 ~ 15:39:07 (7.9s) 15:39:19 ~ 15:39:24 (4.9s) 15:39:31 ~ 15:39:46 (13.6s) 15:40:54 ~ 15:41:04 (7.7s)

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) 1,744 ms 6,553 ms 2,437 ms 10,646 ms 4,346 ms

首文本(SSE delta) 1,745 ms 6,554 ms 2,458 ms 10,647 ms -

总耗时 4,028 ms 6,554 ms 4,092 ms 12,459 ms 5,225 ms

输入 10 0 0 21,628 10

输出 221 219 218 222 230

缓存创建

241

5m: 0 / 1h: 241

5m: 0 / 1h: 0

21,734

5m: 0 / 1h: 21,734

缓存读取 21,545 0 0 0 0

输出速度 96.8 tok/s burst 133.4 tok/s 122.5 tok/s burst

响应

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: haiku
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

步骤 3: cutoff text

时间 15:40:58 ~ 15:41:08 (8.6s) 15:41:10 ~ 15:41:30 (18.0s) 15:40:46 ~ 15:40:56 (8.1s) 15:41:54 ~ 15:42:19 (20.2s) 15:44:42 ~ 15:44:47 (3.9s)

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) 6,626 ms 16,813 ms 2,188 ms 18,789 ms 2,206 ms

首文本(SSE delta) 6,628 ms 16,814 ms 2,190 ms 18,790 ms -

总耗时 7,526 ms 16,815 ms 4,264 ms 19,303 ms 2,781 ms

输入 10 0 0 21,694 10

输出 76 39 259 64 83

缓存创建

261

5m: 0 / 1h: 261

5m: 0 / 1h: 0

5m: 0 / 1h: 61

缓存读取 21,786 0 0 0 21,734

输出速度 84.6 tok/s burst 124.9 tok/s 124.8 tok/s 144.3 tok/s

响应

2025-02

2025-02

Ready. No incomplete work from the previous context. What's next?

2025-02

2025-02

步骤 4: identity_free text

时间 15:42:51 ~ 15:43:16 (21.6s) 15:42:41 ~ 15:42:51 (7.3s) 15:44:24 ~ 15:44:30 (4.3s) 15:43:57 ~ 15:44:02 (4.8s) 15:47:21 ~ 15:47:31 (7.3s)

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) 2,112 ms 6,200 ms 2,146 ms 1,882 ms 2,135 ms

首文本(SSE delta) 2,114 ms 6,201 ms 2,148 ms 1,883 ms -

总耗时 20,697 ms 6,201 ms 3,323 ms 3,794 ms 4,969 ms

输入 10 0 0 21,737 10

输出 2,130 39 118 256 328

缓存创建

118

5m: 0 / 1h: 118

5m: 0 / 1h: 0

21,838

5m: 0 / 1h: 21,838

缓存读取 22,047 0 0 0 0

输出速度 114.6 tok/s burst 100.4 tok/s 134.0 tok/s 115.7 tok/s

响应

我是Claude Haiku，由Anthropic公司开发的AI助手。可协助编程、分析、写作等多种任务，知识截止2025年2月。

我是Claude，来自Anthropic的Haiku模型。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮助编写代码、分析问题、进行研究和完成各类软件工程任务。当前运行的是 Haiku 4.5 模型。

我是Claude，由Anthropic公司开发的AI助手。我擅长软件工程任务，包括代码编写、调试、重构等，能帮助开发者高效完成复杂的编程工作。

我是Claude，由Anthropic公司开发的AI助手。我可以帮助用户完成编程、写作、分析和问题解答等各种任务。

步骤 5: knowledge_recall text

时间 15:44:20 ~ 15:44:45 (20.8s) 15:44:52 ~ 15:45:17 (22.5s) 15:47:25 ~ 15:47:40 (14.6s) 15:48:00 ~ 15:48:15 (13.7s) 15:49:08 ~ 15:49:33 (22.1s)

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) 1,382 ms 21,090 ms 1,862 ms 6,812 ms 2,096 ms

首文本(SSE delta) 1,383 ms 21,091 ms 1,864 ms 6,909 ms -

总耗时 19,926 ms 21,368 ms 13,870 ms 12,619 ms 19,203 ms

输入 10 0 0 9,581 10

输出 2,145 2,078 2,157 1,092 2,196

缓存创建

11,246

5m: 0 / 1h: 11,246

5m: 0 / 1h: 0

8,832

5m: 0 / 1h: 8,832

缓存读取 0 0 0 0 0

输出速度 115.7 tok/s - 179.7 tok/s 191.2 tok/s 128.4 tok/s

响应

```json
{"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

{"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

```json
{"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

```json
{"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

```json
{"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

步骤 6: digit_count text

时间 15:47:27 ~ 15:47:57 (25.5s) 15:48:58 ~ 15:49:18 (15.2s) 15:49:45 ~ 15:49:55 (9.5s) 15:49:29 ~ 15:49:54 (23.4s) 15:50:42 ~ 15:51:12 (25.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=755769926233781815988121915877935698127163827074023222324798627008480398

HTTP 头(TTFB) 1,113 ms 13,867 ms 1,859 ms 10,856 ms 1,021 ms

首文本(SSE delta) 1,115 ms 13,868 ms 1,889 ms 10,857 ms -

总耗时 23,664 ms 14,092 ms 8,710 ms 22,189 ms 24,716 ms

输入 193 0 0 237 193

输出 2,614 1,646 1,162 1,128 4,057

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0 0

输出速度 115.9 tok/s burst 170.4 tok/s 99.5 tok/s 171.2 tok/s

响应

```json
{"n":72,"c":[5,10,8]}
```

{"n":72,"c":[5,10,8]}

```json
{"n":72,"c":[5,10,8]}
```

```json
{"n":72,"c":[5,10,8]}
```

```json
{"n":72,"c":[5,10,8]}
```

总计

总输入(含缓存) 99,032 0 0 96,296 95,987

总输出 7,236 4,043 4,193 2,855 6,961

总缓存创建

18,032

5m: 0 / 1h: 18,032

5m: 0 / 1h: 0

74,010

5m: 0 / 1h: 74,010

总缓存读取 80,757 0 0 0 21,734

输出速度 114.3 tok/s burst 163.2 tok/s 130.3 tok/s 150.6 tok/s

总执行时间 1m 25.2s 1m 16.2s 49.4s 1m 20.9s 1m 10.9s

总墙钟时间

10m 56.2s

含步骤间等待

12m 19.1s

含步骤间等待

13m 2.0s

含步骤间等待

12m 59.1s

含步骤间等待

14m 15.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0670

Anthropic: Claude Haiku 4.5

>=$0.0202

Anthropic: Claude Haiku 4.5 (成本不完整)

>=$0.0210

Anthropic: Claude Haiku 4.5 (成本不完整)

$0.1106

Anthropic: Claude Haiku 4.5

$0.1297

Anthropic: Claude Haiku 4.5

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 (75%) one-api (100%) one-api (100%)

上游 Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (75%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN - - - - -

ID 格式 msg_01SYwipsiHPD212dGCG99TU8 msg_897005ec1c6283e767afeb9f msg_a9d00149f49e4b2c943737e12273e593 msg_2ce130d5f37049929454c0ceae15bd98 msg_018Xun9s7oqhmrjNHvQGFWZS

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: true usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011Ccay7YTNXtUKL6XBiNBap
u-cc	server: nginx x-new-api-version: 1.0.17 x-oneapi-request-id: 202607010748595817053918268d9d6wUv0olJr	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
R-Kiro	server: Photon-Edge x-amzn-requestid: dcdf4933-379d-4a27-a862-9390439b60f2	id_format: msg_prefix has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.cache_fieldsupstream:x-amzn-requestidupstream:anthropic-direct-from-body	1782892186373-ses39ewciia5x dcdf4933-379d-4a27-a862-9390439b60f2
R-Kiro	server: nginx x-new-api-version: v0.0.0 x-oneapi-request-id: 202607010749307189485538IRxg5nW	id_format: msg_prefix has_service_tier: false has_inference_geo: false model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:model:date_suffixbody:usage.cache_fieldsupstream:anthropic-direct-from-body	—
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202607010750435715839098268d9d620esvxad	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—

协议指纹评分（机器自动）

总分 / 100 100 58 51 52 85

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 0.0 /10 ×20 0.0 /10 ×20 0.0 /10 ×20

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

延迟基线 10.0 /10 ×15 5.7 /10 ×15 10.0 /10 ×15 5.0 /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 5.0 /10 ×14 5.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 5.0 /10 ×13 5.0 /10 ×13 5.0 /10 ×13 5.3 /10 ×13

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

系统提示纯净 10.0 /10 ×8 未计入未计入 10.0 /10 ×8 10.0 /10 ×8

身份(结构) 10.0 /10 ×7 6.7 /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 0.0 /10 ×6 0.0 /10 ×6 10.0 /10 ×6

知识截止 10.0 /10 ×5 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

inference_geo 10.0 /10 ×5 10.0 /10 ×5 0.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5

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

Req-ID 透传 10.0 /10 ×2 0.0 /10 ×2 0.0 /10 ×2 0.0 /10 ×2 0.0 /10 ×2

SDK 一致 10.0 /10 ×2 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未计入

综合结论

与基线相似度基准（参考） 58 v3.26.0 51 v3.26.0 52 v3.26.0 85 v3.26.0

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 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 Thinking 体量 thinking_volume_match
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 自报身份一致 self_identity_consistency
20 信封自报一致 envelope_self_report_match
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 13012 ms 11321 ms 6332 ms 12386 ms 9950 ms

输出速度 114.3 tok/s burst 163.2 tok/s 130.3 tok/s 150.6 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 ⓘ

$0.0670

>=$0.0202 (成本不完整)

>=$0.0210 (成本不完整)

$0.1106

$0.1297

缓存读取占比 ⓘ 82% - - 0% 23%

揭晓

通道

官方基准

服务商 HongMaCC ↗

通道u-cc

服务商 SSSAiCode ↗

通道R-Kiro

服务商 NowCoding ↗

通道R-Kiro

服务商 ByteCatCode ↗

通道O-Max

返回列表