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。

3 通道 · 6 步骤

2026/06/17 运行日期

claude-sonnet-4-6 官方基准

3 任务失败执行不完整

已揭晓揭晓进度

⚠ 2 通道不可测 · 已从主对比表移除点击展开原因

DBAI / M-
```
评测失败
```
Aiberm / cc
```
评测失败
```

步骤 / 指标

ByteCatCode

O-Max

SSSAiCode

O-Max

官方基准

基本信息

服务名 Claude Code Claude Code Claude Code

模型 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6

客户端 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 12:37:05 ~ 12:37:10 (4.7s) 12:37:10 ~ 12:37:15 (2.6s) 12:37:16 ~ 12:37:21 (3.9s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,580 ms 1,727 ms 2,875 ms

首文本(SSE delta) - 1,814 ms 2,877 ms

总耗时 2,731 ms 1,872 ms 3,013 ms

输入 3 3 3

输出 5 5 5

缓存创建

32,367

5m: 0 / 1h: 32,367

31,990

5m: 0 / 1h: 31,990

32,318

5m: 0 / 1h: 32,318

缓存读取 0 0 0

响应

pong

pong

pong

步骤 2: identity text

时间 12:41:03 ~ 12:41:08 (4.3s) 12:39:07 ~ 12:39:12 (3.3s) 12:39:16 ~ 12:39:31 (3.6s)

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。严格按以下 3 行格式输出，全部小写英文，不要任何其他文字、解释、标点：
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) 2,160 ms 2,196 ms 2,111 ms

首文本(SSE delta) - 2,477 ms 2,112 ms

总耗时 3,207 ms 2,542 ms 2,430 ms

输入 3 3 3

输出 22 22 22

缓存创建

191

5m: 0 / 1h: 191

191

5m: 0 / 1h: 191

191

5m: 0 / 1h: 191

缓存读取 32,367 31,990 32,318

响应

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-5

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

步骤 3: cutoff text

时间 12:44:11 ~ 12:44:17 (4.3s) 12:41:54 ~ 12:41:59 (3.1s) 12:43:22 ~ 12:43:27 (3.2s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,348 ms 1,673 ms 1,988 ms

首文本(SSE delta) - 1,903 ms 1,990 ms

总耗时 3,203 ms 2,049 ms 2,169 ms

输入 3 3 3

输出 8 8 8

缓存创建

5m: 0 / 1h: 38

5m: 0 / 1h: 57

缓存读取 32,558 32,181 32,509

响应

2025-08

2025-04

2025-08

步骤 4: identity_free text

时间 12:46:55 ~ 12:47:05 (5.2s) 12:43:00 ~ 12:43:05 (2.6s) 12:47:07 ~ 12:47:12 (3.7s)

提示词

查看本步 Prompt

中文回答：你叫什么、属于哪个产品、谁开发？50 字内。

HTTP 头(TTFB) 2,475 ms 1,531 ms 1,668 ms

首文本(SSE delta) - 1,855 ms 1,670 ms

总耗时 3,239 ms 1,918 ms 2,660 ms

输入 3 3 3

输出 48 26 70

缓存创建

5m: 0 / 1h: 41

5m: 0 / 1h: 43

缓存读取 32,596 32,219 32,566

输出速度 62.8 tok/s - 70.7 tok/s

响应

我叫 Claude，是由 Anthropic 公司开发的 AI 助手，当前使用的模型版本为 claude-sonnet-4-6。

我叫 Claude，是由 Anthropic 开发的 AI 助手。

我是 Claude，由 Anthropic 开发的 AI 助手。我能帮助回答问题、分析代码、撰写文档等多种任务。当前运行的模型版本是 claude-sonnet-4-6。

步骤 5: knowledge_recall text

时间 12:49:25 ~ 12:49:36 (5.8s) 12:44:30 ~ 12:44:35 (4.2s) 12:51:04 ~ 12:51:14 (6.0s)

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) 2,040 ms 1,335 ms 2,021 ms

首文本(SSE delta) - 3,432 ms 2,022 ms

总耗时 4,666 ms 3,490 ms 5,051 ms

输入 3 3 3

输出 204 198 205

缓存创建

38,244

5m: 0 / 1h: 38,244

7,675

5m: 0 / 1h: 7,675

8,068

5m: 0 / 1h: 8,068

缓存读取 0 0 0

输出速度 77.7 tok/s - 67.7 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

步骤 6: digit_count text

时间 12:53:09 ~ 12:53:29 (18.4s) 12:46:24 ~ 12:46:29 (0.9s) 12:54:25 ~ 12:54:40 (12.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=335053462650494912397158938261713507636181023758687631349741313007864275

HTTP 头(TTFB) 3,474 ms - 1,178 ms

首文本(SSE delta) - - 1,180 ms

总耗时 16,512 ms - 11,231 ms

输入 3 - 164

输出 1,254 - 756

缓存创建

36,075

5m: 0 / 1h: 36,075

5m: 0 / 1h: 0

缓存读取 0 - 0

输出速度 96.2 tok/s - 75.2 tok/s

响应

Going through `s = 335053462650494912397158938261713507636181023758687631349741313007864275` character by character:

**Position index:**
```
 1: 3   10: 6   19: 3   28: 2   37: 6   46: 7   55: 3   64

失败

评测失败

```json
{"n":72,"c":[6,8,4]}
```

**Verification breakdown** (`s = 335053462650494912397158938261713507636181023758687631349741313007864275`):

| Segment | 0s | 7s | 9s |
|---|---|---|---|
| 335053462

总计

总输入(含缓存)

204,495

136,340

仅前 6 步

138,249

总输出

1,541

259

仅前 6 步

1,066

总缓存创建

106,956

5m: 0 / 1h: 106,956

39,935

5m: 0 / 1h: 39,935

仅前 6 步

40,677

5m: 0 / 1h: 40,677

总缓存读取

97,521

96,390

仅前 6 步

97,393

输出速度

91.7 tok/s

仅前 6 步

73.3 tok/s

总执行时间

42.7s

16.8s

仅前 6 步

32.5s

总墙钟时间

16m 23.6s

含步骤间等待

9m 18.9s

含步骤间等待

仅前 6 步

17m 24.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4535

Anthropic: Claude Sonnet 4.6

$0.1826

Anthropic: Claude Sonnet 4.6 (部分)

$0.1983

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count knowledge_recall digit_count

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

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

CDN - - -

ID 格式 msg_01YZgWcPutsNEDPpXvMRPNVi msg_01LWVz4AbmgTD6mMzGR6X9MN msg_01B7Pv8kcmkrzEWCeBuYtf3u

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

server: nginx/1.29.8
x-oneapi-request-id: 202606170453111408190108268d9d6dX9v4mhV

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

—

O-Max

cf-ray: a0cf6648cf262973-LAX
server: Photon-Edge
set-cookie: _cfuvid=5dt0vLw5UbG3n9FKUINriWST8Bb.hAZUmiqCI9ziVdg-1781671471.4866488-1.0.1.1-a8mJtdsoGSAt7T4Ua8dxC2G3E0detJfadZw34ILqmRc; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com

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

1781671471430-fhvkn5s5c1u30
req_011Cc8DqUqYBETNajyPb6YH3
a0cf6648cf262973-LAX

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_011Cc8EbNLqn2E8TVGzx1kDA

协议指纹评分（机器自动）

总分 / 100 89 97

基线状态已对照基线已对照基线

缓存命中比 3.0 /10 ×20 — 7.3 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 — 10.0 /10 ×15

模型匹配 10.0 /10 ×14 — 10.0 /10 ×14

缓存连续性 10.0 /10 ×14 — 10.0 /10 ×14

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

原生 msg-ID 10.0 /10 ×8 — 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 — 10.0 /10 ×8

身份(结构) 10.0 /10 ×7 — 10.0 /10 ×7

知识截止 10.0 /10 ×7 — 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 — 10.0 /10 ×7

service_tier 10.0 /10 ×6 — 10.0 /10 ×6

inference_geo 10.0 /10 ×5 — 10.0 /10 ×5

延迟基线 10.0 /10 ×5 — 10.0 /10 ×5

Req-ID 透传 0.0 /10 ×4 — 10.0 /10 ×4

stop_reason 10.0 /10 ×3 — 10.0 /10 ×3

SDK 一致 10.0 /10 ×2 — 10.0 /10 ×2

流式投递未计入 — 未计入

综合结论

与基线相似度 89 v3.21.0 执行失败，不评分基准（参考）

雷达图例

共 20 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 自由身份 identity_free_clean
5 停止原因 stop_reason_present
6 系统提示纯净 system_prompt_clean
7 Service Tier service_tier_present
8 缓存命中比 cache_hit_ratio_match
9 缓存 TTL 一致 cache_ttl_consistency
10 推理地区 inference_geo_present
11 缓存连续 cache_continuity_intra
12 知识召回 knowledge_recall_match
13 延迟基线 latency_baseline_match
14 原生 msg-ID anthropic_msg_id_format
15 Sliding 缓存 cache_sliding_correctness
16 结构化身份 identity_structured_match
17 信封自报一致 envelope_self_report_match
18 中文遵循 instruction_following_lang
19 档位思考量 tier_thinking_volume_match
20 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 5593 ms 2374 ms 4426 ms

输出速度 91.7 tok/s - 73.3 tok/s

完成步骤 6/6 成功 5/6 成功 · 1 失败 6/6 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4535

$0.1826 (部分)

$0.1983

缓存读取占比 ⓘ 48% 71% (部分) 70%

揭晓

通道

服务商 ByteCatCode ↗

通道O-Max

服务商 SSSAiCode ↗

通道O-Max

官方基准

返回列表