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

2026/06/30 运行日期

claude-opus-4-8 官方基准

2 任务失败执行不完整

已揭晓揭晓进度

步骤 / 指标

SSSAiCode

O-Max

Aiide

O-Max

SSSAiCode

O-Max-L2

LingxiCode

M-

官方基准

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 23:22:45 ~ 23:22:50 (4.3s) 23:22:43 ~ 23:22:48 (4.9s) 23:22:47 ~ 23:22:52 (3.8s) 23:22:49 ~ 23:22:59 (5.1s) 23:22:51 ~ 23:22:56 (4.6s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,890 ms 3,068 ms 2,353 ms 3,480 ms 3,110 ms

首文本(SSE delta) 2,978 ms 3,069 ms 2,476 ms 3,481 ms 3,111 ms

总耗时 3,052 ms 3,221 ms 2,526 ms 3,870 ms 3,390 ms

输入 2,300 2,326 2,300 2,372 2,326

输出 4 4 4 4 4

缓存创建

1,227

5m: 0 / 1h: 1,227

14,491

5m: 0 / 1h: 14,491

1,229

5m: 0 / 1h: 1,229

1,747

5m: 0 / 1h: 1,713

1,711

5m: 0 / 1h: 1,711

缓存读取 12,732 0 12,732 12,986 12,732

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 23:24:20 ~ 23:24:25 (4.6s) 23:24:25 ~ 23:24:35 (5.2s) 23:26:18 ~ 23:26:23 (4.7s) 23:26:40 ~ 23:26:50 (5.4s) 23:24:01 ~ 23:24:06 (4.7s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 2,134 ms 3,771 ms 2,194 ms 3,707 ms 3,409 ms

首文本(SSE delta) 3,152 ms 3,772 ms 3,692 ms 3,708 ms 3,412 ms

总耗时 3,925 ms 4,188 ms 3,947 ms 4,476 ms 3,717 ms

输入 2 2 2 2 2

输出 73 28 96 28 28

缓存创建

2,492

5m: 0 / 1h: 2,492

2,518

5m: 0 / 1h: 2,518

2,492

5m: 0 / 1h: 2,492

2,568

5m: 0 / 1h: 2,518

2,518

5m: 0 / 1h: 2,518

缓存读取 13,959 14,491 13,961 14,733 14,443

输出速度 94.4 tok/s - - - -

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 23:25:44 ~ 23:25:49 (3.4s) 23:25:37 ~ 23:25:43 (4.9s) 23:27:37 ~ 23:27:42 (3.8s) 23:30:21 ~ 23:30:31 (9.3s) 23:26:16 ~ 23:26:26 (6.4s)

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) 2,272 ms 3,512 ms 2,774 ms 7,941 ms 5,179 ms

首文本(SSE delta) 2,454 ms 3,514 ms 2,941 ms 7,942 ms 5,181 ms

总耗时 2,556 ms 3,514 ms 3,016 ms 8,383 ms 5,454 ms

输入 2 2 2 2 2

输出 6 6 6 6 6

缓存创建

109

5m: 0 / 1h: 109

5m: 0 / 1h: 62

130

5m: 0 / 1h: 130

5m: 0 / 1h: 64

5m: 0 / 1h: 54

缓存读取 16,451 17,009 16,453 17,302 16,961

响应

2025-03

2026-01

2025-01

2026-01

2026-01

步骤 4: identity_free text

时间 23:28:47 ~ 23:28:52 (4.0s) 23:27:14 ~ 23:27:24 (5.9s) 23:29:10 ~ 23:29:15 (4.1s) 23:33:24 ~ 23:33:34 (5.1s) 23:28:18 ~ 23:28:23 (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) 2,315 ms 3,358 ms 2,053 ms 2,876 ms 2,279 ms

首文本(SSE delta) 2,972 ms 3,359 ms 2,955 ms 2,877 ms 2,283 ms

总耗时 3,127 ms 4,607 ms 3,345 ms 4,234 ms 3,015 ms

输入 2 2 2 2 2

输出 49 66 55 62 50

缓存创建

5m: 0 / 1h: 37

5m: 0 / 1h: 39

5m: 0 / 1h: 31

5m: 0 / 1h: 37

缓存读取 16,560 17,071 16,583 17,367 17,015

输出速度 - 52.9 tok/s 141.0 tok/s 45.7 tok/s 68.3 tok/s

响应

我是 Claude，由 Anthropic 开发，属于 Claude 系列 AI 助手中的 Opus 4.8 模型。

我是 Claude，由 Anthropic 开发的 AI 助手。基于 Claude Opus 4.8 模型，可帮助完成软件工程、写作、分析等多种任务。

我是 Claude，一个由 Anthropic 开发的人工智能助手。我可以帮助你完成写作、编程、分析、问答等各类任务。

我是 Claude，一个由 Anthropic 开发的 AI 助手。具体型号是 Opus 4.8，能协助处理编程、写作、分析等多种任务。

我是 Claude，由 Anthropic 开发，属于 Claude 系列 AI 模型，具体型号为 Opus 4.8。

步骤 5: knowledge_recall text

时间 23:31:53 ~ 23:32:03 (6.1s) 23:29:08 ~ 23:29:23 (13.6s) 23:32:46 ~ 23:32:56 (6.2s) 23:34:55 ~ 23:35:10 (10.3s) 23:32:14 ~ 23:32:24 (9.5s)

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) 2,178 ms 2,776 ms 2,115 ms 2,553 ms 2,431 ms

首文本(SSE delta) 2,784 ms 5,265 ms 2,672 ms 4,033 ms 3,966 ms

总耗时 5,449 ms 9,993 ms 5,459 ms 9,464 ms 8,457 ms

输入 2 2 2 2 2

输出 347 688 353 731 688

缓存创建

5,283

5m: 0 / 1h: 5,283

5,765

5m: 0 / 1h: 5,765

5,306

5m: 0 / 1h: 5,306

5,872

5m: 0 / 1h: 5,757

5,739

5m: 0 / 1h: 5,739

缓存读取 0 0 0 0 0

输出速度 130.2 tok/s 145.5 tok/s 126.7 tok/s 134.6 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 23:33:22 ~ 23:33:27 (1.2s) 23:32:44 ~ 23:32:54 (5.1s) 23:34:30 ~ 23:34:35 (1.2s) 23:38:37 ~ 23:38:42 (4.8s) 23:34:07 ~ 23:34:12 (3.1s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=714963002705917192250021307050943409628024525960696308013070399940200275

HTTP 头(TTFB) - 2,670 ms - 3,240 ms 1,807 ms

首文本(SSE delta) - 2,671 ms - 3,252 ms 1,809 ms

总耗时 - 3,124 ms - 3,943 ms 2,188 ms

输入 - 211 - 215 211

输出 - 19 - 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 - 0 - 0 0

响应

失败

评测失败

{"n":72,"c":[14,6,9]}

失败

评测失败

{"n":72,"c":[14,5,9]}

{"n":72,"c":[12,7,9]}

总计

总输入(含缓存)

71,158

仅前 6 步

73,991

71,225

仅前 6 步

75,266

73,755

总输出

479

仅前 6 步

811

514

仅前 6 步

850

795

总缓存创建

9,148

5m: 0 / 1h: 9,148

仅前 6 步

22,875

5m: 0 / 1h: 22,875

9,188

5m: 0 / 1h: 9,188

仅前 6 步

10,283

5m: 0 / 1h: 10,083

10,059

5m: 0 / 1h: 10,059

总缓存读取

59,702

仅前 6 步

48,571

59,729

仅前 6 步

62,388

61,151

输出速度

122.2 tok/s

仅前 6 步

126.2 tok/s

128.4 tok/s

仅前 6 步

116.8 tok/s

68.3 tok/s

总执行时间

23.6s

仅前 6 步

39.5s

23.8s

仅前 6 步

40.0s

32.3s

总墙钟时间

10m 42.0s

含步骤间等待

仅前 6 步

10m 11.0s

含步骤间等待

11m 48.2s

含步骤间等待

仅前 6 步

15m 53.4s

含步骤间等待

11m 20.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1105

Anthropic: Claude Opus 4.8 (部分)

$0.2003

Anthropic: Claude Opus 4.8

$0.1117

Anthropic: Claude Opus 4.8 (部分)

$0.1284

Anthropic: Claude Opus 4.8

$0.1260

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 knowledge_recall digit_count knowledge_recall digit_count digit_count

平台 direct (100%) one-api (100%) direct (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_01WjarVLYMgdLuDFNsD3aRni msg_01X5yR2CePn1HjGAw8uQpFf9 msg_014P1EURYw6h2qiVjoS5P5yg msg_01Rs5GtUvhy1ddYpffYpqine msg_01UNLDiPztdBWoeRpAzc2Z8p

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: a13e387a1fb5cba7-LAX server: Photon-Edge set-cookie: _cfuvid=dRj9TqcVsA3PpvM3uOI0Q4.6xdz9sakFGzvMmtGWzXI-1782833514.588521-1.0.1.1-sLOd.FnXo1AY9HUtEqelZN._camNcgs72P_Im6D1ck8; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com	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	1782833514471-vuzu1sv4dple8 req_011CcZgigR6gtD5vSEwbLKob a13e387a1fb5cba7-LAX
O-Max	cf-ray: a13e39bebaa40e90-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606301532466039866848268d9d686P85PhD	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	a13e39bebaa40e90-AMS
O-Max-L2	cf-ray: a13e39c6dce74711-LAX server: Photon-Edge set-cookie: _cfuvid=B3Kf_Gypz86wHeDEdOzu0K0KKVJABDuKo6NjhX9RH7g-1782833567.8156552-1.0.1.1-rVVaCY6VD0SsaFWBQbjTBkVUIdmy.mh8CcVuJvkNxPU; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com	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	1782833567695-oe7bmpe1kmqhj req_011CcZgnbntytV1XfawbBePv a13e39c6dce74711-LAX
M-	cf-ray: a13e4258ea207343-NRT server: cloudflare set-cookie: server_name_session=7cb5035642cad6865e35ea3751ae010f; Max-Age=86400; httponly; path=/ x-oneapi-request-id: 2026063015383960856549324BZeNny	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	a13e4258ea207343-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_011CcZgtZF2Wiixq3YWJkVAj

协议指纹评分（机器自动）

总分 / 100 92 97 100

基线状态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

缓存连续性 — 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

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

缓存命中比 — 9.0 /10 ×20 — 10.0 /10 ×20 10.0 /10 ×20

身份(结构) — 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

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

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

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

stop_reason — 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

inference_geo — 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

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

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

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

综合结论

与基线相似度执行失败，不评分 92 v3.26.0 执行失败，不评分 97 v3.26.0 基准（参考）

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 3622 ms 4775 ms 3659 ms 5728 ms 4370 ms

输出速度 122.2 tok/s 126.2 tok/s 128.4 tok/s 116.8 tok/s 68.3 tok/s

完成步骤 5/6 成功 · 1 失败 6/6 成功 5/6 成功 · 1 失败 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 ⓘ

$0.1105 (部分)

$0.2003

$0.1117 (部分)

$0.1284

$0.1260

缓存读取占比 ⓘ 84% (部分) 66% 84% (部分) 83% 83%

揭晓

通道

服务商 SSSAiCode ↗

通道O-Max

服务商 Aiide ↗

通道O-Max

服务商 SSSAiCode ↗

通道O-Max-L2

服务商 LingxiCode ↗

通道M-

官方基准

返回列表