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

2026/06/25 运行日期

claude-opus-4-8 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

Aiide

O-Max

官方基准

luminai

M-Kiro/Antigravity

ByteCatCode

O-Max

SAIAi

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 v2.1.160 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 12:18:26 ~ 12:18:36 (5.1s) 12:18:30 ~ 12:18:35 (3.2s) 12:18:24 ~ 12:18:34 (6.5s) 12:18:28 ~ 12:18:38 (8.2s) 12:18:22 ~ 12:18:27 (3.9s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,501 ms 1,867 ms 3,963 ms 5,127 ms 2,483 ms

首文本(SSE delta) 3,503 ms 1,869 ms 3,965 ms - 2,484 ms

总耗时 3,649 ms 1,992 ms 4,023 ms 5,318 ms 2,628 ms

输入 1,454 1,454 1,643 1,454 1,454

输出 4 4 1 4 4

缓存创建

30,380

5m: 0 / 1h: 30,380

5m: 0 / 1h: 0

566

5m: 566 / 1h: 0

30,218

5m: 0 / 1h: 30,218

30,222

5m: 0 / 1h: 30,222

缓存读取 0 30,222 18,318 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 12:20:33 ~ 12:20:43 (6.5s) 12:21:38 ~ 12:21:43 (3.6s) 12:21:48 ~ 12:21:58 (6.4s) 12:21:35 ~ 12:21:40 (4.3s) 12:20:05 ~ 12:20:10 (4.8s)

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) 3,285 ms 1,848 ms 4,886 ms 2,717 ms 3,417 ms

首文本(SSE delta) 3,286 ms 1,850 ms 4,888 ms - 3,419 ms

总耗时 3,739 ms 2,222 ms 4,978 ms 3,171 ms 3,790 ms

输入 2 2 1,657 2 2

输出 28 28 23 28 28

缓存创建

1,634

5m: 0 / 1h: 1,634

5m: 0 / 1h: 0

571

5m: 571 / 1h: 0

1,634

5m: 0 / 1h: 1,634

1,634

5m: 0 / 1h: 1,634

缓存读取 30,380 31,856 18,478 30,218 30,222

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 12:23:41 ~ 12:23:46 (4.6s) 12:25:33 ~ 12:25:38 (3.3s) 12:25:57 ~ 12:26:07 (5.0s) 12:22:44 ~ 12:22:49 (4.0s) 12:21:41 ~ 12:21:46 (3.2s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,042 ms 1,868 ms 3,706 ms 2,635 ms 1,937 ms

首文本(SSE delta) 3,081 ms 1,870 ms 3,708 ms - 1,938 ms

总耗时 3,267 ms 2,347 ms 3,941 ms 2,896 ms 2,155 ms

输入 2 2 1,662 2 2

输出 6 6 4 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 64

573

5m: 573 / 1h: 0

5m: 0 / 1h: 54

缓存读取 32,014 31,856 18,531 31,852 31,856

响应

2026-01

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 12:26:28 ~ 12:26:38 (9.0s) 12:29:33 ~ 12:29:38 (4.6s) 12:27:12 ~ 12:27:22 (7.3s) 12:24:30 ~ 12:24:40 (5.1s) 12:23:15 ~ 12:23:20 (4.0s)

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) 5,141 ms 2,784 ms 5,555 ms 2,772 ms 2,018 ms

首文本(SSE delta) 5,181 ms 2,787 ms 5,973 ms - 2,019 ms

总耗时 6,545 ms 3,423 ms 5,973 ms 3,925 ms 3,044 ms

输入 2 2 1,665 2 2

输出 51 48 28 60 61

缓存创建

5m: 0 / 1h: 37

574

5m: 574 / 1h: 0

5m: 0 / 1h: 31

5m: 0 / 1h: 35

缓存读取 32,068 31,920 18,570 31,906 31,910

输出速度 37.4 tok/s 75.5 tok/s - 52.0 tok/s 59.5 tok/s

响应

我是 Claude，属于 Claude 产品系列，由 Anthropic 公司开发。具体模型为 Claude Opus 4.8。

我是 Claude，属于 Claude 产品系列，由 Anthropic 公司开发。具体模型为 Opus 4.8。

我是 Claude Opus 4.8，属于 Claude 产品系列，由 Anthropic 开发。

我是 Claude，由 Anthropic 开发的人工智能助手，具体为 Claude Opus 4.8 模型，可协助完成软件工程等各类任务。

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、写作、分析等多种任务。

步骤 5: knowledge_recall text

时间 12:29:25 ~ 12:29:35 (6.6s) 12:31:16 ~ 12:31:26 (9.2s) 12:29:21 ~ 12:29:36 (13.8s) 12:28:18 ~ 12:28:33 (11.5s) 12:27:03 ~ 12:27:13 (8.7s)

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) 2,389 ms 2,634 ms 7,040 ms 2,872 ms 2,390 ms

首文本(SSE delta) 2,390 ms 6,074 ms 7,041 ms - 5,286 ms

总耗时 5,289 ms 8,178 ms 12,770 ms 8,861 ms 7,699 ms

输入 2 2 268 2 2

输出 325 677 385 652 720

缓存创建

4,752

5m: 0 / 1h: 4,752

2,987

5m: 0 / 1h: 2,987

5m: 92 / 1h: 0

4,751

5m: 0 / 1h: 4,751

4,760

5m: 0 / 1h: 4,760

缓存读取 0 1,772 2,989 0 0

输出速度 112.1 tok/s burst 67.2 tok/s 108.9 tok/s 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

{"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 12:32:22 ~ 12:32:27 (3.7s) 12:35:08 ~ 12:35:13 (2.6s) 12:32:36 ~ 12:33:01 (20.4s) 12:31:56 ~ 12:32:26 (25.6s) 12:29:14 ~ 12:29:19 (3.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=647022071396156269157141560962599600293889956852177412139722442887083401

HTTP 头(TTFB) 2,184 ms 1,269 ms 5,427 ms 10,796 ms 1,690 ms

首文本(SSE delta) 2,185 ms 1,270 ms 5,428 ms - 1,692 ms

总耗时 2,693 ms 1,602 ms 19,253 ms 24,262 ms 2,187 ms

输入 211 211 16 762 211

输出 19 19 110 141 19

缓存创建

5m: 0 / 1h: 0

5m: 5 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 168 0 0

输出速度 - - 8.0 tok/s 10.5 tok/s -

响应

{"n":72,"c":[6,9,11]}

{"n":72,"c":[6,9,11]}

{"n":72,"c":[7,7,9]}

{"n":72,"c":[7,7,9]}

{"n":72,"c":[6,8,9]}

总计

总输入(含缓存) 132,992 132,387 86,346 132,888 132,366

总输出 433 782 551 891 838

总缓存创建

36,857

5m: 0 / 1h: 36,857

3,088

5m: 0 / 1h: 3,088

2,381

5m: 2,381 / 1h: 0

36,688

5m: 0 / 1h: 36,688

36,705

5m: 0 / 1h: 36,705

总缓存读取 94,462 127,626 77,054 93,976 93,988

输出速度 88.2 tok/s 75.5 tok/s 25.3 tok/s 41.4 tok/s 59.5 tok/s

总执行时间 35.4s 26.5s 59.4s 58.6s 27.8s

总墙钟时间

14m 0.9s

含步骤间等待

16m 42.9s

含步骤间等待

14m 37.0s

含步骤间等待

13m 58.0s

含步骤间等待

10m 57.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2968

Anthropic: Claude Opus 4.8

$0.1110

Anthropic: Claude Opus 4.8

$0.0962

Anthropic: Claude Opus 4.8

$0.3097

Anthropic: Claude Opus 4.8

$0.3057

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count digit_count

平台 one-api (100%) direct (100%) one-api (100%) one-api (100%) direct (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_01AN9Eu8V2XFcj6H9sdBx1Zh msg_01WBgJQqU1uNwdqMDYyMGuru msg_01VEydK2JS146LbRqcFkwPZCQ2 msg_01HQDhVFvLrEarCkaL7V45zy msg_018JcKve6uVs1jcuDPQ7wEFh

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	cf-ray: a1113f82e8bb1c86-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606250432237407315368268d9d6Sav814oc	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a1113f82e8bb1c86-AMS
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_011CcPMbPGH67d7kvCGrbQXF
M-Kiro/Antigravity	cf-ray: a1113fdb9a9f655d-AMS server: cloudflare x-oneapi-request-id: 202606250432378054888138268d9d6cTMzDWcb	id_format: msg_prefix extra_fields: ["container"] has_service_tier: true has_inference_geo: false 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.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a1113fdb9a9f655d-AMS
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202606250431579725579638268d9d6bgOZswZ5	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_011CcPM9JGRBeBndNPPK7Tya

协议指纹评分（机器自动）

总分 / 100 95 100 74 93 100

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 10.0 /10 ×20

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

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

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

综合结论

与基线相似度 95 v3.24.1 基准（参考） 74 v3.24.1 93 v3.24.1 100 v3.24.1

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 信封自报一致 envelope_self_report_match
20 中文遵循 instruction_following_lang
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 4197 ms 3294 ms 8490 ms 8072 ms 3584 ms

输出速度 88.2 tok/s 75.5 tok/s 25.3 tok/s 41.4 tok/s 59.5 tok/s

完成步骤 6/6 成功 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 ⓘ

$0.2968

$0.1110

$0.0962

$0.3097

$0.3057

缓存读取占比 ⓘ 71% 96% 89% 71% 71%

揭晓

通道

服务商 Aiide ↗

通道O-Max

官方基准

服务商 luminai ↗

通道M-Kiro/Antigravity

服务商 ByteCatCode ↗

通道O-Max

服务商 SAIAi ↗

通道O-Max

返回列表