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

2026/07/02 运行日期

claude-haiku-4-5-20251001 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

TopRouterCN / O-Max
```
评测失败
```

步骤 / 指标

官方基准

AnyAIGC

M-

SAIAi

O-Max

ByteCatCode

O-Max

基本信息

服务名 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

客户端 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 12:28:25 ~ 12:28:30 (3.2s) 12:28:21 ~ 12:28:26 (4.2s) 12:28:23 ~ 12:28:28 (4.1s) 12:28:17 ~ 12:28:22 (4.8s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 1,865 ms 2,862 ms 3,086 ms 2,011 ms

首文本(SSE delta) 1,867 ms 2,866 ms 3,087 ms -

总耗时 2,207 ms 2,866 ms 3,150 ms 2,745 ms

输入 10 9,391 10 10

输出 53 2 46 55

缓存创建

6,166

5m: 0 / 1h: 6,166

6,166

5m: 0 / 1h: 6,166

21,545

5m: 0 / 1h: 21,545

缓存读取 15,379 0 15,379 0

输出速度 155.9 tok/s - - 74.9 tok/s

响应

pong

pong

pong

pong

步骤 2: identity text

时间 12:32:24 ~ 12:32:34 (5.0s) 12:31:40 ~ 12:31:50 (7.3s) 12:32:06 ~ 12:32:16 (6.5s) 12:32:14 ~ 12:32:19 (4.9s)

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) 2,651 ms 5,196 ms 3,042 ms 1,814 ms

首文本(SSE delta) 2,653 ms 5,197 ms 3,043 ms -

总耗时 4,160 ms 5,808 ms 5,548 ms 3,830 ms

输入 10 9,576 10 10

输出 187 25 248 198

缓存创建

233

5m: 0 / 1h: 233

226

5m: 0 / 1h: 226

178

5m: 0 / 1h: 178

缓存读取 21,545 0 21,545 21,545

输出速度 124.1 tok/s - 99.0 tok/s 98.2 tok/s

响应

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

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

步骤 3: cutoff text

时间 12:35:08 ~ 12:35:13 (3.0s) 12:35:31 ~ 12:35:36 (4.2s) 12:33:29 ~ 12:33:34 (3.5s) 12:34:57 ~ 12:35:02 (3.9s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 1,508 ms 2,914 ms 1,761 ms 1,943 ms

首文本(SSE delta) 1,510 ms 2,915 ms 1,763 ms -

总耗时 2,014 ms 2,915 ms 2,483 ms 2,844 ms

输入 10 9,637 10 10

输出 70 5 67 91

缓存创建

210

5m: 0 / 1h: 210

290

5m: 0 / 1h: 290

5m: 0 / 1h: 61

缓存读取 21,778 0 21,771 21,723

输出速度 138.9 tok/s - 93.1 tok/s 101.0 tok/s

响应

2025-02

2025-05

2025-02

2025-02

步骤 4: identity_free text

时间 12:37:24 ~ 12:37:34 (9.2s) 12:37:26 ~ 12:37:31 (4.5s) 12:36:53 ~ 12:37:08 (10.1s) 12:38:11 ~ 12:38:21 (7.4s)

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) 2,168 ms 3,332 ms 3,342 ms 2,377 ms

首文本(SSE delta) 2,170 ms 3,333 ms 3,343 ms -

总耗时 8,178 ms 3,343 ms 9,126 ms 6,403 ms

输入 10 9,673 10 10

输出 835 28 608 460

缓存创建

112

5m: 0 / 1h: 112

101

5m: 0 / 1h: 101

5m: 0 / 1h: 43

缓存读取 21,988 0 22,061 21,784

输出速度 139.0 tok/s - 105.1 tok/s 114.3 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。擅长编程、分析、写作等任务。当前运行的是 Claude Haiku 4.5 模型。

我叫Claude，是由Anthropic开发的AI助手，属于Claude产品系列，旨在提供智能协助和信息服务。

我是Claude，由Anthropic开发的AI助手。版本为Haiku 4.5，知识截止2025年2月。

我是Claude Code，由Anthropic公司开发的AI编程助手。我可以帮助用户进行代码编写、调试、测试和代码审查等软件工程任务。

步骤 5: knowledge_recall text

时间 12:38:43 ~ 12:38:58 (13.3s) 12:40:58 ~ 12:41:08 (6.8s) 12:39:10 ~ 12:39:55 (42.8s) 12:42:09 ~ 12:42:34 (20.6s)

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) 2,345 ms 3,849 ms 1,532 ms 1,883 ms

首文本(SSE delta) 2,346 ms 3,850 ms 1,533 ms -

总耗时 12,379 ms 4,585 ms 41,945 ms 19,509 ms

输入 10 10,302 10 10

输出 1,802 154 4,727 2,082

缓存创建

9,886

5m: 0 / 1h: 9,886

9,721

5m: 0 / 1h: 9,721

8,828

5m: 0 / 1h: 8,828

缓存读取 0 0 0 0

输出速度 179.6 tok/s 209.5 tok/s 117.0 tok/s 118.1 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":"UNKNOWN","oscars2024_best_picture":"UNKNOWN","ucl_2024":"UNKNOWN","nba_2024":"UNKNOWN","copa_america_2024":"UNKNOWN","euro2024":"UNKNOWN","nobel_peace_2024":"UNKNOWN","wo

```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 12:41:02 ~ 12:41:22 (16.7s) 12:42:58 ~ 12:43:03 (4.9s) 12:42:26 ~ 12:43:16 (47.4s) 12:44:21 ~ 12:44:46 (24.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=463150385736482234175782296253762057217082056444833739872912125758444476

HTTP 头(TTFB) 1,841 ms 2,051 ms 924 ms 2,244 ms

首文本(SSE delta) 1,842 ms 2,052 ms 925 ms -

总耗时 15,734 ms 3,648 ms 46,460 ms 23,100 ms

输入 193 165 193 193

输出 2,105 14 5,701 2,391

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 151.5 tok/s - 125.2 tok/s 114.6 tok/s

响应

```json
{"n":72,"c":[4,10,3]}
```

{"n":81,"c":[6,8,3]}

```json
{"n": 72, "c": [4, 10, 3]}
```

```json
{"n":72,"c":[4,10,3]}
```

总计

总输入(含缓存) 97,540 48,744 97,503 95,950

总输出 5,052 228 11,397 5,277

总缓存创建

16,607

5m: 0 / 1h: 16,607

16,504

5m: 0 / 1h: 16,504

30,655

5m: 0 / 1h: 30,655

总缓存读取 80,690 0 80,756 65,052

输出速度 156.5 tok/s 209.5 tok/s 119.5 tok/s 114.3 tok/s

总执行时间 50.5s 31.8s 1m 54.5s 1m 5.8s

总墙钟时间

12m 57.0s

含步骤间等待

14m 41.9s

含步骤间等待

14m 53.1s

含步骤间等待

16m 29.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0543

Anthropic: Claude Haiku 4.5

$0.0499

Anthropic: Claude Haiku 4.5

$0.0859

Anthropic: Claude Haiku 4.5

$0.0715

Anthropic: Claude Haiku 4.5

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 direct (100%) direct (50%) direct (100%) one-api (100%)

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

CDN - - - -

ID 格式 msg_019jPhU9Fexs5qu7sdY2g4Yq 自定义 msg_01C9pGu3sHvBqRL17EWUY5kx msg_016CdKYjPjRLA5S1aTy4yZic

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_011CccchtV9ccQkNCrjQXcvA
M-	server: nginx	id_format: non_standard has_service_tier: false has_inference_geo: false model_has_date_suffix: true usage_has_cache_fields: true	body:id:non_standardbody:model:date_suffixbody:usage.cache_fields	202607020442599612496198268d9d6gc3Wxo7a
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: 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_011Ccccp5bZ4SwkKAfkMms1z
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202607020444226266561498268d9d6C3B22JBy	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 99 37 98 99

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 0.0 /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

延迟基线 10.0 /10 ×15 9.2 /10 ×15 9.9 /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 5.0 /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

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

系统提示纯净 10.0 /10 ×8 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

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 0.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

知识截止 10.0 /10 ×5 5.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

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

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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度基准（参考） 37 v3.26.0 98 v3.26.0 99 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 7445 ms 3861 ms 18119 ms 9739 ms

输出速度 156.5 tok/s 209.5 tok/s 119.5 tok/s 114.3 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 ⓘ

$0.0543

$0.0499

$0.0859

$0.0715

缓存读取占比 ⓘ 83% 0% 83% 68%

揭晓

通道

官方基准

服务商 AnyAIGC ↗

通道M-

服务商 SAIAi ↗

通道O-Max

服务商 ByteCatCode ↗

通道O-Max

返回列表