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。

4 通道 · 5 步骤

2026/06/16 运行日期

claude-opus-4-8 官方基准

1 任务失败执行不完整

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

SAIAi

O-Max

官方基准

FastCode

ClaudeCN

O-Max

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8

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 04:38:51 ~ 04:38:56 (3.5s) 04:38:53 ~ 04:38:58 (3.2s) 04:38:49 ~ 04:38:59 (5.9s) 04:38:47 ~ 04:38:57 (9.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,258 ms 1,910 ms 3,903 ms 7,714 ms

首文本(SSE delta) 2,260 ms 1,912 ms 3,904 ms 7,716 ms

总耗时 2,412 ms 2,067 ms 4,533 ms 7,927 ms

输入 1,454 2 1,454 1,454

输出 4 4 4 4

缓存创建

30,222

5m: 0 / 1h: 30,222

1,452

5m: 1,452 / 1h: 0

30,222

5m: 0 / 1h: 30,222

30,225

5m: 0 / 1h: 30,225

缓存读取 0 30,222 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 04:42:23 ~ 04:42:28 (3.5s) 04:41:09 ~ 04:41:14 (3.5s) 04:42:31 ~ 04:42:41 (5.4s) 04:42:21 ~ 04:42:31 (5.4s)

提示词

查看本步 Prompt

输出 3 行 metadata 描述你自己（当前 API 模型）。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。禁止任何额外字符：
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) 1,951 ms 2,077 ms 3,601 ms 3,412 ms

首文本(SSE delta) 1,952 ms 2,079 ms 3,605 ms 3,413 ms

总耗时 2,380 ms 2,486 ms 3,885 ms 3,914 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

5m: 0 / 1h: 0

1,643

5m: 0 / 1h: 1,643

1,643

5m: 0 / 1h: 1,643

1,643

5m: 0 / 1h: 1,643

缓存读取 31,865 30,222 30,222 30,225

响应

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

步骤 3: cutoff text

时间 04:46:04 ~ 04:46:14 (7.1s) 04:44:03 ~ 04:44:08 (4.7s) 04:44:22 ~ 04:44:32 (7.4s) 04:45:02 ~ 04:45:12 (6.0s)

提示词

查看本步 Prompt

你的 knowledge cutoff 是？请只回 YYYY-MM。

HTTP 头(TTFB) 5,746 ms 3,417 ms 4,549 ms 4,540 ms

首文本(SSE delta) 5,757 ms 3,418 ms 4,556 ms 4,541 ms

总耗时 6,014 ms 3,719 ms 6,204 ms 4,802 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 56

5m: 0 / 1h: 54

5m: 0 / 1h: 71

5m: 0 / 1h: 62

缓存读取 31,865 31,865 31,865 31,868

响应

2026-01

2025-01

2026-01

2026-01

步骤 4: identity_free text

时间 04:48:43 ~ 04:48:48 (3.8s) 04:47:15 ~ 04:47:20 (3.9s) 04:47:35 ~ 04:47:45 (5.7s) 04:47:11 ~ 04:47:21 (5.0s)

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) 2,100 ms 2,262 ms 3,336 ms 3,296 ms

首文本(SSE delta) 2,102 ms 2,263 ms 3,337 ms 3,298 ms

总耗时 2,807 ms 2,887 ms 4,484 ms 3,876 ms

输入 2 2 2 2

输出 49 46 61 44

缓存创建

5m: 0 / 1h: 37

5m: 0 / 1h: 31

5m: 0 / 1h: 35

缓存读取 31,921 31,919 31,936 31,930

输出速度 69.5 tok/s 73.7 tok/s 53.2 tok/s 76.1 tok/s

响应

我是 Claude，具体型号为 Claude Opus 4.8，属于 Claude 产品系列，由 Anthropic 开发。

我是 Claude，一个 AI 助手，由 Anthropic 开发。当前模型是 Claude Opus 4.8。

我是 Claude，由 Anthropic 打造的 AI 助手，当前模型为 Claude Opus 4.8，可协助编程、研究、写作等多种任务。

我是 Claude，由 Anthropic 打造的 AI 助手，当前模型为 Claude Opus 4.8。

步骤 5: knowledge_recall text

时间 04:50:51 ~ 04:51:01 (9.0s) 04:49:21 ~ 04:49:31 (8.1s) 04:50:54 ~ 04:51:04 (9.3s) 04:48:41 ~ 04:48:56 (10.2s)

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) 2,377 ms 2,321 ms 2,903 ms 4,027 ms

首文本(SSE delta) 5,653 ms 4,640 ms 5,844 ms 6,771 ms

总耗时 8,027 ms 7,098 ms 8,145 ms 9,120 ms

输入 2 2 2 2

输出 693 595 674 686

缓存创建

2,989

5m: 0 / 1h: 2,989

4,750

5m: 0 / 1h: 4,750

4,786

5m: 0 / 1h: 4,786

4,763

5m: 0 / 1h: 4,763

缓存读取 1,772 0 0 0

输出速度 burst burst burst burst

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

总计

总输入(含缓存) 132,189 132,168 132,242 132,213

总输出 780 679 773 768

总缓存创建

33,304

5m: 0 / 1h: 33,304

7,930

5m: 1,452 / 1h: 6,478

36,757

5m: 0 / 1h: 36,757

36,728

5m: 0 / 1h: 36,728

总缓存读取 97,423 124,228 94,023 94,023

输出速度 69.5 tok/s 73.7 tok/s 53.2 tok/s 76.1 tok/s

总执行时间 26.8s 23.4s 33.7s 36.1s

总墙钟时间

12m 9.9s

含步骤间等待

10m 37.8s

含步骤间等待

12m 14.8s

含步骤间等待

10m 8.8s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2837

Anthropic: Claude Opus 4.8

$0.1253

Anthropic: Claude Opus 4.8

$0.3034

Anthropic: Claude Opus 4.8

$0.3031

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall

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

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

CDN - - - cloudflare

ID 格式 msg_01A17gjwo9cnuckGhSMLGHej msg_01QzgCvADvYYJA42sHjqc2rq msg_012Kr27WiBmXHMs3DQaYGRvB msg_01B1pc9V462eyoZH47XHj2Fa

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/1.22.1	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_011Cc5hufvPpHz85iL5vBeyJ
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_011Cc5ho2L314ZAo9jeiu1Wd
cc	server: nginx	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	—
O-Max	cf-ray: a0c46fe7a824669f-AMS server: cloudflare x-oneapi-request-id: 20260616044842483388233t5Ia4TsX	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	a0c46fe7a824669f-AMS

协议指纹评分（机器自动）

总分 / 100 100 90 97 97

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

缓存 TTL 一致 10.0 /10 ×15 7.0 /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 10.0 /10 ×13 10.0 /10 ×13 10.0 /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 10.0 /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 0.0 /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 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 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

Req-ID 透传 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

SDK 一致 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 100 v3.16.0 基准（参考） 97 v3.16.0 97 v3.16.0

雷达图例

共 19 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 缓存连续 cache_continuity_intra
13 知识召回 knowledge_recall_match
14 延迟基线 latency_baseline_match
15 原生 msg-ID anthropic_msg_id_format
16 Sliding 缓存 cache_sliding_correctness
17 结构化身份 identity_structured_match
18 中文遵循 instruction_following_lang
19 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 4328 ms 3651 ms 5450 ms 5928 ms

输出速度 69.5 tok/s 73.7 tok/s 53.2 tok/s 76.1 tok/s

完成步骤 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 ⓘ

$0.2837

$0.1253

$0.3034

$0.3031

缓存读取占比 ⓘ 74% 94% 71% 71%

揭晓

通道

服务商 SAIAi ↗

通道O-Max

官方基准

服务商 FastCode ↗

通道cc

服务商 ClaudeCN ↗

通道O-Max

返回列表