模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

同一会话连续 5 步，覆盖 5 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。步与步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；全流程总跨度 > 6 分钟容纳新增的第 5 步。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 · 5 步骤

2026/06/16 运行日期

claude-sonnet-4-6 官方基准

1 通道分数 < 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

RunAPI

O-Max

SAIAi

O-Max

官方基准

LingxiCode

M-

FoxCode

R-ultra

基本信息

服务名 Claude Code 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 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 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 11:45:58 ~ 11:46:08 (5.6s) 11:45:54 ~ 11:45:59 (4.3s) 11:46:02 ~ 11:46:07 (4.0s) 11:45:56 ~ 11:46:01 (4.6s) 11:46:00 ~ 11:46:05 (3.6s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,189 ms 2,946 ms 2,783 ms 3,199 ms 2,101 ms

首文本(SSE delta) 3,192 ms 2,948 ms 2,785 ms 3,201 ms 2,102 ms

总耗时 4,186 ms 3,089 ms 2,905 ms 3,357 ms 2,669 ms

输入 3 3 3 3 61

输出 5 5 5 5 14

缓存创建

32,320

5m: 0 / 1h: 32,320

32,318

5m: 0 / 1h: 32,318

5m: 0 / 1h: 0

32,318

5m: 0 / 1h: 32,318

32,955

5m: 32,955 / 1h: 0

缓存读取 0 0 32,318 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 11:47:20 ~ 11:47:25 (3.6s) 11:47:18 ~ 11:47:23 (3.3s) 11:48:34 ~ 11:48:44 (5.9s) 11:49:43 ~ 11:49:48 (4.7s) 11:48:36 ~ 11:48:46 (5.5s)

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) 1,839 ms 1,977 ms 4,378 ms 3,164 ms 4,074 ms

首文本(SSE delta) 1,840 ms 1,979 ms 4,380 ms 3,165 ms 4,076 ms

总耗时 2,285 ms 2,315 ms 4,963 ms 3,554 ms 4,608 ms

输入 3 3 3 3 16

输出 22 22 22 22 65

缓存创建

191

5m: 0 / 1h: 191

191

5m: 0 / 1h: 191

5m: 0 / 1h: 0

191

5m: 0 / 1h: 191

558

5m: 558 / 1h: 0

缓存读取 32,320 32,318 32,509 32,318 32,955

输出速度 - - - - 122.2 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

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

步骤 3: cutoff text

时间 11:49:57 ~ 11:50:07 (5.0s) 11:49:54 ~ 11:49:59 (3.8s) 11:52:32 ~ 11:52:37 (4.3s) 11:51:24 ~ 11:51:29 (4.8s) 11:52:10 ~ 11:52:15 (4.9s)

提示词

查看本步 Prompt

你的训练数据知识截止到哪一年哪一月？只用 YYYY-MM 格式回答，不要解释。

HTTP 头(TTFB) 3,398 ms 2,474 ms 3,080 ms 3,552 ms 3,315 ms

首文本(SSE delta) 3,409 ms 2,609 ms 3,162 ms 3,553 ms 3,316 ms

总耗时 3,884 ms 2,801 ms 3,230 ms 3,749 ms 3,967 ms

输入 3 3 3 3 62

输出 8 8 8 8 32

缓存创建

5m: 0 / 1h: 64

5m: 0 / 1h: 44

5m: 0 / 1h: 38

5m: 0 / 1h: 45

33,857

5m: 33,857 / 1h: 0

缓存读取 32,511 32,509 32,509 32,509 0

输出速度 - - - - 49.2 tok/s

响应

2025-08

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 11:51:55 ~ 11:52:00 (3.7s) 11:53:31 ~ 11:53:36 (4.7s) 11:54:30 ~ 11:54:35 (4.2s) 11:54:02 ~ 11:54:07 (4.2s) 11:54:26 ~ 11:54:36 (5.1s)

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) 1,793 ms 2,953 ms 2,132 ms 2,765 ms 3,187 ms

首文本(SSE delta) 1,795 ms 2,954 ms 2,136 ms 2,767 ms 3,188 ms

总耗时 2,554 ms 3,764 ms 3,175 ms 3,551 ms 4,197 ms

输入 3 3 3 3 9

输出 49 55 62 57 79

缓存创建

5m: 0 / 1h: 43

5m: 0 / 1h: 35

5m: 0 / 1h: 39

5m: 0 / 1h: 35

422

5m: 422 / 1h: 0

缓存读取 32,575 32,553 32,547 32,554 33,857

输出速度 64.6 tok/s 67.9 tok/s 59.7 tok/s 72.7 tok/s 78.3 tok/s

响应

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。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于大型语言模型构建，旨在提供安全、有益的对话和任务协助。

我是 Claude，由 Anthropic 公司打造的 AI 助手。当前版本为 claude-sonnet-4-6，致力于提供安全、有帮助的对话与任务协助。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于大型语言模型构建，旨在提供安全、有帮助的对话和任务协助。

我是 Kiro，由 Anthropic 构建的 AI 助手，基于 Claude 模型（claude-sonnet-4-6），帮助开发者完成软件工程及各类专业任务。

步骤 5: knowledge_recall text

时间 11:54:39 ~ 11:54:50 (5.5s) 11:55:53 ~ 11:55:58 (4.4s) 11:57:35 ~ 11:57:41 (4.4s) 11:56:27 ~ 11:56:37 (6.4s) 11:57:15 ~ 11:57:36 (17.3s)

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) 1,421 ms 1,467 ms 1,573 ms 2,860 ms 2,192 ms

首文本(SSE delta) 1,422 ms 1,468 ms 1,574 ms 2,861 ms 2,193 ms

总耗时 4,448 ms 3,383 ms 3,392 ms 5,479 ms 16,495 ms

输入 3 3 3 3 55

输出 205 204 205 205 1,342

缓存创建

8,056

5m: 0 / 1h: 8,056

8,032

5m: 0 / 1h: 8,032

2,319

5m: 0 / 1h: 2,319

8,035

5m: 0 / 1h: 8,035

9,619

5m: 9,619 / 1h: 0

缓存读取 0 0 5,718 0 0

输出速度 67.7 tok/s 106.5 tok/s 112.8 tok/s 78.3 tok/s 93.8 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

{"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

总计

总输入(含缓存) 138,095 138,015 138,012 138,020 144,426

总输出 289 294 302 297 1,532

总缓存创建

40,674

5m: 0 / 1h: 40,674

40,620

5m: 0 / 1h: 40,620

2,396

5m: 0 / 1h: 2,396

40,624

5m: 0 / 1h: 40,624

77,411

5m: 77,411 / 1h: 0

总缓存读取 97,406 97,380 135,601 97,381 66,812

输出速度 67.1 tok/s 95.0 tok/s 93.5 tok/s 77.0 tok/s 92.0 tok/s

总执行时间 23.4s 20.5s 22.9s 24.8s 36.3s

总墙钟时间

8m 51.7s

含步骤间等待

10m 4.8s

含步骤间等待

11m 38.8s

含步骤间等待

10m 40.9s

含步骤间等待

11m 35.8s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1861

Anthropic: Claude Sonnet 4.6

$0.1860

Anthropic: Claude Sonnet 4.6

$0.0542

Anthropic: Claude Sonnet 4.6

$0.1861

Anthropic: Claude Sonnet 4.6

$0.2251

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall knowledge_recall

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

CDN - - - cloudflare cloudflare

ID 格式 msg_0182cAcULEg5tYAjLrXkG2YP msg_015C7hZK8NuGN7jqZhLaV5bV msg_01YEBNxYbv6gu6mprTS5y7XN msg_0199NqGFqspqBzQmiusfLYLx msg_01SbQKfDRwKZKo4N07hAfG6E

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/1.24.0 (Ubuntu) x-oneapi-request-id: 20260616115441219022628aEhv9Uii	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	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_011Cc6GKb6L7bJwmBwofHT7t
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_011Cc6GT81dPHJWBkWYpVave
M-	cf-ray: a0c6e2833afee393-NRT server: cloudflare set-cookie: server_name_session=3de609b8cf2a7e28d141c3b6c8b41f57; Max-Age=86400; httponly; path=/ x-oneapi-request-id: 20260616035628760602989XEU9ASR0	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	a0c6e2833afee393-NRT
R-ultra	cf-ray: a0c6e3b3ed34e3bf-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606160357169421825148268d9d6lYgXz1zY	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	a0c6e3b3ed34e3bf-NRT

协议指纹评分（机器自动）

总分 / 100 97 100 100 97 74

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 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 5.2 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 0.0 /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 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 10.0 /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 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

系统提示纯净 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 5.0 /10 ×8

身份(结构) 10.0 /10 ×7 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 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 0.0 /10 ×7

service_tier 10.0 /10 ×6 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 10.0 /10 ×5

延迟基线 10.0 /10 ×5 10.0 /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 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 10.0 /10 ×3

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 — — — — —

综合结论

与基线相似度 97 v3.17.0 100 v3.17.0 基准（参考） 97 v3.17.0 74 v3.17.0

雷达图例

共 18 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 中文遵循 instruction_following_lang
18 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 3471 ms 3070 ms 3533 ms 3938 ms 6387 ms

输出速度 67.1 tok/s 95.0 tok/s 93.5 tok/s 77.0 tok/s 92.0 tok/s

完成步骤 5/5 成功 5/5 成功 5/5 成功 5/5 成功 5/5 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1861

$0.1860

$0.0542

$0.1861

$0.2251

缓存读取占比 ⓘ 71% 71% 98% 71% 46%

揭晓

通道

服务商 RunAPI ↗

通道O-Max

服务商 SAIAi ↗

通道O-Max

官方基准

服务商 LingxiCode ↗

通道M-

服务商 FoxCode ↗

通道R-ultra

返回列表