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

2026/06/28 运行日期

claude-sonnet-4-6 官方基准

3 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

InfAI / R-Kiro
```
评测失败
```

步骤 / 指标

FoxCode

R-super

官方基准

FastCode

PackyCode

cc-Zeus

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-sonnet-4-6 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 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 23:09:32 ~ 23:09:37 (3.5s) 23:09:38 ~ 23:09:43 (4.8s) 23:09:36 ~ 23:09:46 (4.9s) 23:09:34 ~ 23:09:39 (4.7s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,032 ms 3,420 ms 3,496 ms 2,578 ms

首文本(SSE delta) 2,033 ms 3,422 ms 3,498 ms 2,579 ms

总耗时 2,490 ms 3,569 ms 3,498 ms 2,580 ms

输入 30 3 3 3

输出 15 5 5 5

缓存创建

32,986

5m: 32,986 / 1h: 0

32,318

5m: 0 / 1h: 32,318

32,318

5m: 0 / 1h: 32,318

32,320

5m: 0 / 1h: 32,320

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 23:12:39 ~ 23:12:49 (6.4s) 23:13:25 ~ 23:13:30 (4.0s) 23:11:14 ~ 23:11:24 (7.7s) 23:13:14 ~ 23:13:19 (4.6s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 3,685 ms 2,459 ms 5,681 ms 2,349 ms

首文本(SSE delta) 3,686 ms 2,461 ms 5,682 ms 2,350 ms

总耗时 5,470 ms 3,016 ms 5,682 ms 3,172 ms

输入 79 3 3 3

输出 140 22 22 22

缓存创建

33,458

5m: 33,458 / 1h: 0

192

5m: 0 / 1h: 192

192

5m: 0 / 1h: 192

32,510

5m: 0 / 1h: 32,510

缓存读取 0 32,318 32,318 0

输出速度 78.5 tok/s - - -

响应

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

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

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

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

步骤 3: cutoff text

时间 23:15:29 ~ 23:15:34 (4.4s) 23:14:33 ~ 23:14:38 (3.3s) 23:14:59 ~ 23:15:09 (6.3s) 23:14:22 ~ 23:14:32 (5.1s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,093 ms 2,001 ms 4,567 ms 2,962 ms

首文本(SSE delta) 3,094 ms 2,172 ms 4,568 ms 2,963 ms

总耗时 3,419 ms 2,326 ms 4,597 ms 3,177 ms

输入 33 3 3 3

输出 51 8 8 8

缓存创建

506

5m: 506 / 1h: 0

5m: 0 / 1h: 57

5m: 0 / 1h: 55

缓存读取 33,458 32,510 32,510 32,510

响应

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 23:18:29 ~ 23:18:39 (5.1s) 23:17:46 ~ 23:17:51 (3.3s) 23:17:48 ~ 23:17:58 (6.9s) 23:16:42 ~ 23:16:47 (4.5s)

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) 3,223 ms 1,805 ms 4,595 ms 1,819 ms

首文本(SSE delta) 3,224 ms 1,807 ms 4,596 ms 1,820 ms

总耗时 4,146 ms 2,333 ms 5,351 ms 3,004 ms

输入 78 3 3 3

输出 90 41 55 43

缓存创建

34,307

5m: 34,307 / 1h: 0

5m: 0 / 1h: 39

5m: 0 / 1h: 43

5m: 0 / 1h: 41

缓存读取 0 32,567 32,567 32,565

输出速度 97.6 tok/s 77.9 tok/s 72.8 tok/s 36.3 tok/s

响应

我是 Kiro，由 AWS 打造的 AI 编程助手，底层模型为Anthropic 的 Claude（claude-sonnet-4-6）。

我是 Claude，由 Anthropic 公司打造的 AI 助手。当前版本为 claude-sonnet-4-6。

我是 Claude，由 Anthropic 开发的 AI 助手。我能帮助你解答问题、编写代码、分析文本、进行创作等多种任务。

我叫 Claude，是由 Anthropic 公司开发的 AI 助手。当前版本为 Claude Sonnet 4.6。

步骤 5: knowledge_recall text

时间 23:20:11 ~ 23:20:36 (24.8s) 23:20:32 ~ 23:20:37 (4.3s) 23:20:13 ~ 23:20:23 (8.1s) 23:20:09 ~ 23:20:19 (5.4s)

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) 1,424 ms 1,377 ms 3,764 ms 1,390 ms

首文本(SSE delta) 1,425 ms 1,379 ms 3,765 ms 1,391 ms

总耗时 23,926 ms 3,242 ms 6,641 ms 4,351 ms

输入 78 3 3 3

输出 2,086 205 205 204

缓存创建

9,701

5m: 9,701 / 1h: 0

8,036

5m: 0 / 1h: 8,036

8,054

5m: 0 / 1h: 8,054

8,038

5m: 0 / 1h: 8,038

缓存读取 0 0 0 0

输出速度 92.7 tok/s 110.0 tok/s 71.3 tok/s 68.9 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 23:24:17 ~ 23:24:42 (22.3s) 23:22:48 ~ 23:23:13 (20.9s) 23:21:38 ~ 23:21:53 (12.8s) 23:24:15 ~ 23:24:35 (18.4s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=792658024872241353780362639404603549951138393100243627526890545941370746

HTTP 头(TTFB) 1,920 ms 1,215 ms 4,167 ms 2,383 ms

首文本(SSE delta) 1,921 ms 1,608 ms 4,168 ms 2,384 ms

总耗时 21,395 ms 19,944 ms 10,709 ms 17,379 ms

输入 12 164 164 164

输出 2,379 1,559 832 1,262

缓存创建

197

5m: 197 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 122.2 tok/s 85.0 tok/s 127.2 tok/s 84.2 tok/s

响应

{"n":72,"c":[8,6,7]}

```
s = 792658024872241353780362639404603549951138393100243627526890545941370746
```

**Digit-by-digit breakdown (grouped in 10s):**

| Group | Digits |
|-------|--------|
| 1–10 | 7 9 2 6 5 8 **0** 2

```json
{"n":72,"c":[8,6,7]}
```

**Verification:**

String: `792658024872241353780362639404603549951138393100243627526890545941370746`

- **Length (n):** 72 digits
- **0s:** positions 7,21,29,32,47,4

## Counting digits in s

**s** = `792658024872241353780362639404603549951138393100243627526890545941370746`

### Step 1 — Total length

Splitting into groups of 10:

| Group | Digits |
|-------|------

总计

总输入(含缓存) 144,923 138,216 138,238 138,218

总输出 4,761 1,840 1,127 1,544

总缓存创建

111,155

5m: 111,155 / 1h: 0

40,642

5m: 0 / 1h: 40,642

40,664

5m: 0 / 1h: 40,664

72,964

5m: 0 / 1h: 72,964

总缓存读取 33,458 97,395 97,395 65,075

输出速度 105.1 tok/s 87.1 tok/s 107.4 tok/s 78.8 tok/s

总执行时间 1m 6.5s 40.6s 46.7s 42.6s

总墙钟时间

15m 10.0s

含步骤间等待

13m 34.9s

含步骤间等待

12m 17.0s

含步骤间等待

15m 1.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.3429

Anthropic: Claude Sonnet 4.6

$0.2098

Anthropic: Claude Sonnet 4.6

$0.1992

Anthropic: Claude Sonnet 4.6

$0.3168

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN cloudflare - - cloudflare

ID 格式 msg_01DBhOBrV16ZrwKyVIfbKSTo msg_018F1dbRQ6rJN7yJkD9nW9Lq msg_bdrk_01R4pBPogogpdGbxY45D39gU msg_01JUX9Q4p9r8ak7ERbKatfbz

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-super	cf-ray: a12db2991d16e3bf-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606281524189540704838268d9d6CB8NKggK	id_format: msg_prefix 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_fieldsupstream:anthropic-direct-from-body	a12db2991d16e3bf-NRT
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_011CcVtQgfrd76gqTr5sf32b
cc	server: nginx	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
cc-Zeus	via: 1.1 Caddy cf-ray: a12db28dd9402f40-LAX server: cloudflare x-oneapi-request-id: 20260628232416959519648TzvxDaJe	id_format: msg_prefix via_chain: ["1.1 Caddy"] 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-rayhdr:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a12db28dd9402f40-LAX

协议指纹评分（机器自动）

总分 / 100 68 98 82 84

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 0.0 /10 ×20 8.3 /10 ×20 8.3 /10 ×20 0.1 /10 ×20

缓存 TTL 一致 0.0 /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 5.3 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 7.7 /10 ×13

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

系统提示纯净 5.0 /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) 0.0 /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 10.0 /10 ×6

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

延迟基线 9.5 /10 ×5 10.0 /10 ×5 5.9 /10 ×5 10.0 /10 ×5

流式投递 9.1 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

Req-ID 透传 0.0 /10 ×4 10.0 /10 ×4 0.0 /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

综合结论

与基线相似度 68 v3.25.0 基准（参考） 82 v3.25.0 84 v3.25.0

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 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 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 10141 ms 5738 ms 6080 ms 5611 ms

输出速度 105.1 tok/s 87.1 tok/s 107.4 tok/s 78.8 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 ⓘ

$0.3429

$0.2098

$0.1992

$0.3168

缓存读取占比 ⓘ 23% 70% 70% 47%

揭晓

通道

服务商 FoxCode ↗

通道R-super

官方基准

服务商 FastCode ↗

通道cc

服务商 PackyCode ↗

通道cc-Zeus

返回列表