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

2026/06/25 运行日期

claude-sonnet-4-6 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

DBAI / M-
```
评测失败
```

步骤 / 指标

FoxCode

R-ultra

官方基准

ClaudeCN

O-Max

OmniaKey

O-Max

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 06:47:23 ~ 06:47:28 (4.0s) 06:47:25 ~ 06:47:30 (4.2s) 06:47:19 ~ 06:47:29 (5.2s) 06:47:21 ~ 06:47:32 (9.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,442 ms 3,021 ms 3,613 ms 7,482 ms

首文本(SSE delta) 2,444 ms 3,025 ms 3,615 ms 7,534 ms

总耗时 2,838 ms 3,150 ms 3,825 ms 7,827 ms

输入 98 3 3 3

输出 15 5 5 5

缓存创建

32,918

5m: 32,918 / 1h: 0

32,318

5m: 0 / 1h: 32,318

32,319

5m: 0 / 1h: 32,319

32,388

5m: 0 / 1h: 32,388

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 06:50:16 ~ 06:51:51 (93.7s) 06:49:23 ~ 06:49:28 (4.6s) 06:49:36 ~ 06:49:46 (5.4s) 06:50:58 ~ 06:51:08 (5.0s)

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) 91,156 ms 3,074 ms 3,632 ms 3,298 ms

首文本(SSE delta) 91,157 ms 3,076 ms 3,633 ms 3,299 ms

总耗时 92,817 ms 3,619 ms 4,051 ms 3,803 ms

输入 52 3 3 3

输出 80 22 22 22

缓存创建

33,472

5m: 33,472 / 1h: 0

191

5m: 0 / 1h: 191

191

5m: 0 / 1h: 191

191

5m: 0 / 1h: 191

缓存读取 0 32,318 32,319 32,388

输出速度 48.2 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 06:55:05 ~ 06:55:15 (6.2s) 06:50:29 ~ 06:50:34 (3.3s) 06:53:33 ~ 06:53:43 (4.9s) 06:54:30 ~ 06:54:35 (4.7s)

提示词

查看本步 Prompt

请直接回答你的知识截止日期，格式 YYYY-MM，不要附加任何说明。

HTTP 头(TTFB) 5,056 ms 1,969 ms 3,557 ms 3,266 ms

首文本(SSE delta) 5,057 ms 2,092 ms 3,558 ms 3,412 ms

总耗时 5,397 ms 2,162 ms 3,840 ms 3,499 ms

输入 53 3 3 3

输出 51 8 8 8

缓存创建

33,897

5m: 33,897 / 1h: 0

5m: 0 / 1h: 57

32,564

5m: 0 / 1h: 32,564

5m: 0 / 1h: 57

缓存读取 0 32,509 0 32,579

输出速度 150.0 tok/s - - -

响应

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 06:58:21 ~ 06:58:31 (5.3s) 06:54:08 ~ 06:54:18 (5.2s) 06:55:26 ~ 06:55:36 (6.3s) 06:57:59 ~ 06:58:14 (10.3s)

提示词

查看本步 Prompt

用中文回答：你是谁？由谁开发？50 字以内。

HTTP 头(TTFB) 2,333 ms 3,304 ms 4,627 ms 9,061 ms

首文本(SSE delta) 2,334 ms 3,305 ms 4,628 ms 9,062 ms

总耗时 4,439 ms 4,001 ms 5,183 ms 9,292 ms

输入 100 3 3 3

输出 125 49 46 19

缓存创建

34,239

5m: 34,239 / 1h: 0

5m: 0 / 1h: 43

5m: 0 / 1h: 41

5m: 0 / 1h: 35

缓存读取 0 32,566 32,564 32,636

输出速度 59.4 tok/s 70.4 tok/s 82.9 tok/s -

响应

我是 Kiro，一个 AI 编程助手，基于 Anthropic 的 Claude 模型（claude-sonnet-4-6）构建，帮助用户完成软件工程及其他专业任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我能帮助你完成写作、分析、编程、问答等各类任务。

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

我是Claude，由Anthropic开发的AI助手。

步骤 5: knowledge_recall text

时间 07:02:11 ~ 07:02:31 (16.8s) 06:56:15 ~ 06:56:20 (4.8s) 06:57:23 ~ 06:57:33 (5.6s) 07:00:17 ~ 07:00:27 (5.5s)

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) 1,578 ms 1,944 ms 2,028 ms 2,439 ms

首文本(SSE delta) 1,579 ms 1,946 ms 2,029 ms 2,493 ms

总耗时 15,946 ms 3,791 ms 4,514 ms 4,290 ms

输入 45 3 3 3

输出 1,236 204 205 198

缓存创建

9,714

5m: 9,714 / 1h: 0

8,047

5m: 0 / 1h: 8,047

8,040

5m: 0 / 1h: 8,040

8,009

5m: 0 / 1h: 8,009

缓存读取 0 0 0 0

输出速度 86.0 tok/s 110.6 tok/s 82.5 tok/s 110.2 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 07:05:37 ~ 07:06:02 (22.9s) 06:59:55 ~ 07:00:20 (20.4s) 06:59:26 ~ 06:59:46 (19.1s) 07:03:35 ~ 07:03:56 (19.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=392026036935470179571626377930882799640588724373595511940706369315705311

HTTP 头(TTFB) 1,960 ms 1,517 ms 2,022 ms 2,501 ms

首文本(SSE delta) 1,961 ms 1,520 ms 2,023 ms 4,043 ms

总耗时 22,013 ms 19,444 ms 18,065 ms 17,968 ms

输入 22 164 164 164

输出 2,765 1,507 1,403 1,554

缓存创建

187

5m: 187 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 137.9 tok/s 84.1 tok/s 87.5 tok/s 111.6 tok/s

响应

{"n":72,"c":[8,10,9]}

Parsing `s = "392026036935470179571626377930882799640588724373595511940706369315705311"`:

**Length count (groups of 10):**
```
3926063693  →  10
5470179571  →  10
6263779308  →  10
8279964058  →  10

## Counting digits in:
`392026036935470179571626377930882799640588724373595511940706369315705311`

**Length:** Grouping in 10s → 7 groups of 10 + 2 = **72**

**Digit-by-digit scan (1-indexed):**

| Di

**Counting the string:**

`s = 392026036935470179571626377930882799640588724373595511940706369315705311`

**Length:** Split into groups of 10:
`3920260369 | 3547017957 | 1626377930 | 8827996405 | 8872

总计

总输入(含缓存) 144,797 138,228 138,217 138,462

总输出 4,272 1,795 1,689 1,806

总缓存创建

144,427

5m: 144,427 / 1h: 0

40,656

5m: 0 / 1h: 40,656

73,155

5m: 0 / 1h: 73,155

40,680

5m: 0 / 1h: 40,680

总缓存读取 0 97,393 64,883 97,603

输出速度 110.5 tok/s 86.0 tok/s 86.7 tok/s 111.4 tok/s

总执行时间 2m 28.8s 42.4s 46.5s 54.2s

总墙钟时间

18m 38.2s

含步骤间等待

12m 55.0s

含步骤间等待

12m 27.0s

含步骤间等待

16m 34.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4037

Anthropic: Claude Sonnet 4.6

$0.2091

Anthropic: Claude Sonnet 4.6

$0.3197

Anthropic: Claude Sonnet 4.6

$0.2095

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

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

CDN cloudflare - cloudflare cloudflare

ID 格式 msg_018MJCz8hUL6L00W78psDvsS msg_01G9utoN4q2BGcm7PN9tcjA7 msg_01V1gSvrRQde1BDz3ZY9G3mD msg_01Mw1ZyB5GcR4RnnBiy6jqY6

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-ultra	cf-ray: a10f60dbed4ae041-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 20260624230538740336608268d9d6JvxLImVP	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	a10f60dbed4ae041-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_011CcNv2nonzFdJre4zaR3oj
O-Max	cf-ray: a10f57d3db5f2908-AMS server: cloudflare x-oneapi-request-id: 20260625065928542342958z0HR790z	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a10f57d3db5f2908-AMS
O-Max	cf-ray: a10f5de9e924bcc9-AMS server: cloudflare x-oneapi-request-id: 202606242303377962604238268d9d6QRn1CwYB	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a10f5de9e924bcc9-AMS

协议指纹评分（机器自动）

总分 / 100 60 100 91 96

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 10.0 /10 ×20 5.5 /10 ×20 10.0 /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 0.0 /10 ×13 10.0 /10 ×13 7.7 /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 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 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 9.7 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 9.2 /10 ×5

流式投递 8.3 /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 60 v3.24.1 基准（参考） 91 v3.24.1 96 v3.24.1

雷达图例

共 23 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 中文遵循 instruction_following_lang
22 档位思考量 tier_thinking_volume_match
23 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 23908 ms 6028 ms 6580 ms 7780 ms

输出速度 110.5 tok/s 86.0 tok/s 86.7 tok/s 111.4 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 ⓘ

$0.4037

$0.2091

$0.3197

$0.2095

缓存读取占比 ⓘ 0% 70% 47% 70%

揭晓

通道

服务商 FoxCode ↗

通道R-ultra

官方基准

服务商 ClaudeCN ↗

通道O-Max

服务商 OmniaKey ↗

通道O-Max

返回列表