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。

5 通道 · 6 步骤

2026/07/01 运行日期

claude-opus-4-8 官方基准

1 通道分数 < 70 另含 1 任务失败

已揭晓揭晓进度

步骤 / 指标

SudoCode

M-Max

官方基准

HongMaCC

u-cc

OmniaKey

O-Max

ByteCatCode

O-Max

基本信息

服务名 Claude Code 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 claude-opus-4-8

客户端 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195 claude-cli v2.1.195

偏差标红基于官方基准

步骤 1: ping text

时间 00:29:23 ~ 00:30:48 (83.7s) 00:29:27 ~ 00:29:32 (3.5s) 00:29:19 ~ 00:29:24 (3.5s) 00:29:25 ~ 00:29:35 (5.0s) 00:29:21 ~ 00:29:31 (5.3s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 5,948 ms 2,108 ms 2,152 ms 3,182 ms 2,518 ms

首文本(SSE delta) 5,950 ms 2,110 ms 2,153 ms 3,183 ms -

总耗时 11,934 ms 2,466 ms 2,153 ms 3,364 ms 2,691 ms

输入 2,326 2,326 0 2,326 2,326

输出 4 4 22 4 4

缓存创建

1,711

5m: 0 / 1h: 1,711

1,711

5m: 0 / 1h: 1,711

5m: 0 / 1h: 0

1,711

5m: 0 / 1h: 1,711

14,449

5m: 0 / 1h: 14,449

缓存读取 12,732 12,732 0 12,732 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 00:33:15 ~ 00:33:30 (12.4s) 00:33:06 ~ 00:33:11 (3.8s) 00:31:33 ~ 00:31:38 (4.0s) 00:31:35 ~ 00:31:45 (5.3s) 00:32:11 ~ 00:32:21 (6.7s)

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。请按下列 3 行格式精确填空作答（label 不变，值小写）：
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) 5,095 ms 2,272 ms 2,875 ms 3,693 ms 2,870 ms

首文本(SSE delta) 5,101 ms 2,273 ms 2,876 ms 3,694 ms -

总耗时 11,443 ms 2,758 ms 2,876 ms 4,224 ms 3,536 ms

输入 2 2 0 2 2

输出 28 28 107 28 28

缓存创建

2,506

5m: 0 / 1h: 2,506

2,506

5m: 0 / 1h: 2,506

5m: 0 / 1h: 0

2,506

5m: 0 / 1h: 2,506

16,955

5m: 0 / 1h: 16,955

缓存读取 14,443 14,443 0 14,443 0

输出速度 - - burst - -

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

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

步骤 3: cutoff text

时间 00:37:08 ~ 00:37:23 (12.7s) 00:35:10 ~ 00:35:15 (3.7s) 00:33:53 ~ 00:33:58 (4.8s) 00:35:26 ~ 00:35:36 (5.6s) 00:35:16 ~ 00:35:26 (9.4s)

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) 5,450 ms 2,556 ms 3,688 ms 2,923 ms 6,295 ms

首文本(SSE delta) 5,451 ms 2,557 ms 3,689 ms 2,924 ms -

总耗时 11,739 ms 2,811 ms 3,689 ms 4,489 ms 8,204 ms

输入 2 2 0 2 2

输出 6 6 41 6 6

缓存创建

5m: 0 / 1h: 64

5m: 0 / 1h: 62

5m: 0 / 1h: 0

5m: 0 / 1h: 62

5m: 0 / 1h: 64

缓存读取 16,949 16,949 0 16,949 16,955

输出速度 - - burst - -

响应

2026-01

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 00:41:20 ~ 00:41:30 (7.2s) 00:38:32 ~ 00:38:37 (3.6s) 00:36:47 ~ 00:36:52 (3.4s) 00:38:43 ~ 00:38:58 (12.8s) 00:38:17 ~ 00:38:22 (4.5s)

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) - 1,877 ms 2,273 ms 5,422 ms 2,879 ms

首文本(SSE delta) - 1,878 ms 2,274 ms 5,423 ms -

总耗时 - 2,661 ms 2,274 ms 7,394 ms 3,480 ms

输入 - 2 0 2 2

输出 - 47 43 54 46

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 0

5m: 0 / 1h: 35

5m: 0 / 1h: 31

缓存读取 - 17,011 0 17,011 17,019

输出速度 - 60.0 tok/s burst 27.4 tok/s 76.5 tok/s

响应

失败

评测失败

我是 Claude，由 Anthropic 开发的 AI 助手，可帮助处理编程、写作、分析等各类任务。

我是Claude，Anthropic开发的AI助手。

我是 Claude，由 Anthropic 公司打造的 AI 助手，当前模型为 Opus 4.8，可协助编程及多种任务。

我是 Claude，由 Anthropic 公司开发的 AI 助手。具体型号是 Claude Opus 4.8。

步骤 5: knowledge_recall text

时间 - 00:41:26 ~ 00:41:41 (11.9s) 00:40:32 ~ 00:40:47 (13.4s) 00:40:30 ~ 00:40:46 (9.9s) 00:42:01 ~ 00:42:16 (11.0s)

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) - 4,467 ms 12,121 ms 4,364 ms 3,825 ms

首文本(SSE delta) - 5,514 ms 12,122 ms 4,365 ms -

总耗时 - 10,851 ms 12,288 ms 8,885 ms 9,870 ms

输入 - 2 0 2 2

输出 - 700 1,145 698 692

缓存创建

5,734

5m: 0 / 1h: 5,734

5m: 0 / 1h: 0

5,737

5m: 0 / 1h: 5,737

5,728

5m: 0 / 1h: 5,728

缓存读取 - 0 0 0 0

输出速度 - 131.2 tok/s burst burst 114.5 tok/s

响应

未执行

{"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 - 00:43:45 ~ 00:43:51 (3.4s) 00:41:55 ~ 00:42:05 (5.1s) 00:42:09 ~ 00:42:14 (4.2s) 00:43:22 ~ 00:43:27 (4.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=610669431433131666235044205432673523674192985161097971379936735161190914

HTTP 头(TTFB) - 1,985 ms 3,899 ms 2,701 ms 2,798 ms

首文本(SSE delta) - 1,986 ms 3,901 ms 2,702 ms -

总耗时 - 2,445 ms 4,009 ms 3,117 ms 3,799 ms

输入 - 211 0 211 211

输出 - 19 703 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 - 0 0 0 0

输出速度 - - burst - -

响应

未执行

{"n":71,"c":[3,8,10]}

{"n":78,"c":[5,6,8]}

{"n":71,"c":[3,7,9]}

{"n":71,"c":[3,7,9]}

总计

总输入(含缓存)

50,735

仅前 4 步

73,732

73,731

73,746

总输出

仅前 4 步

804

2,061

809

795

总缓存创建

4,281

5m: 0 / 1h: 4,281

仅前 4 步

10,052

5m: 0 / 1h: 10,052

5m: 0 / 1h: 0

10,051

5m: 0 / 1h: 10,051

37,227

5m: 0 / 1h: 37,227

总缓存读取

44,124

仅前 4 步

61,135

33,974

输出速度

仅前 4 步

122.1 tok/s

burst

27.4 tok/s

111.0 tok/s

总执行时间

1m 56.1s

仅前 4 步

30.0s

34.2s

42.9s

41.7s

总墙钟时间

12m 7.8s

含步骤间等待

仅前 4 步

14m 23.9s

含步骤间等待

12m 46.1s

含步骤间等待

12m 49.1s

含步骤间等待

14m 6.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0614

Anthropic: Claude Opus 4.8 (部分)

$0.1262

Anthropic: Claude Opus 4.8

>=$0.0515

Anthropic: Claude Opus 4.8 (成本不完整)

$0.1263

Anthropic: Claude Opus 4.8

$0.2823

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 cutoff digit_count digit_count digit_count digit_count

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

ID 格式 msg_01FqssUE9QKv7MUWtQ1rYYwG msg_01GNgJknUz51uzuuztHdSe2s msg_5a78f3c1d5d5a0a54f4943ad msg_019hrv3Uu38tbQM83UZToU6d msg_01LoZkzgGTCT4Zpf7U9VGd6R

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
M-Max	server: nginx x-new-api-version: v1.0.0-rc.10 x-oneapi-request-id: 20260701003709939515219K7KPc1I0	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-new-api-versionhdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
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_011CcZnCaP4odH1untDqn6B1
u-cc	server: nginx x-new-api-version: 1.0.17 x-oneapi-request-id: 202606301641562495629928268d9d6mn6JfSHA	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-new-api-versionhdr: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: a13e9f66b8369deb-AMS server: cloudflare x-oneapi-request-id: 202606301642108382316928268d9d6JCoi8mFM	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	a13e9f66b8369deb-AMS
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202606301643231412974438268d9d6TJqpjUVo	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	—

协议指纹评分（机器自动）

总分 / 100 100 54 99 85

基线状态已对照基线已对照基线已对照基线已对照基线

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 5.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m — 10.0 /10 ×13 5.0 /10 ×13 10.0 /10 ×13 7.7 /10 ×13

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

缓存命中比 — 10.0 /10 ×20 0.0 /10 ×20 10.0 /10 ×20 1.9 /10 ×20

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 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

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

Req-ID 透传 — 10.0 /10 ×2 0.0 /10 ×2 0.0 /10 ×2 0.0 /10 ×2

stop_reason — 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3

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

SDK 一致 — 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2

系统提示纯净 — 10.0 /10 ×8 未计入 10.0 /10 ×8 10.0 /10 ×8

延迟基线 — 10.0 /10 ×15 10.0 /10 ×15 9.6 /10 ×15 9.9 /10 ×15

流式投递 — 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5 未计入

综合结论

与基线相似度执行失败，不评分基准（参考） 54 v3.26.0 99 v3.26.0 85 v3.26.0

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 Thinking 体量 thinking_volume_match
14 缓存连续 cache_continuity_intra
15 知识召回 knowledge_recall_match
16 延迟基线 latency_baseline_match
17 原生 msg-ID anthropic_msg_id_format
18 Sliding 缓存 cache_sliding_correctness
19 结构化身份 identity_structured_match
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 11705 ms 3999 ms 4548 ms 5246 ms 5263 ms

输出速度 - 122.1 tok/s burst 27.4 tok/s 111.0 tok/s

完成步骤 3/6 成功 · 1 失败 · 2 跳过 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 ⓘ

$0.0614 (部分)

$0.1262

>=$0.0515 (成本不完整)

$0.1263

$0.2823

缓存读取占比 ⓘ 87% (部分) 83% - 83% 46%

揭晓

通道

服务商 SudoCode ↗

通道M-Max

官方基准

服务商 HongMaCC ↗

通道u-cc

服务商 OmniaKey ↗

通道O-Max

服务商 ByteCatCode ↗

通道O-Max

返回列表