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

2026/06/15 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

FoxCode

官方基准

luminai

M-Kiro/Antigravity

RunAPI

O-Max

Novaris

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 16:30:32 ~ 16:30:42 (7.3s) 16:30:36 ~ 16:30:41 (3.4s) 16:30:34 ~ 16:30:44 (6.1s) 16:30:30 ~ 16:30:35 (3.6s) 16:30:28 ~ 16:30:38 (6.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 5,927 ms 2,178 ms 4,514 ms 1,755 ms 3,456 ms

首文本(SSE delta) 5,928 ms 2,182 ms 4,515 ms 1,759 ms 3,457 ms

总耗时 6,261 ms 2,330 ms 4,756 ms 2,187 ms 3,674 ms

输入 1,454 1,454 2,053 1,454 1,454

输出 4 4 1 4 4

缓存创建

30,222

5m: 30,222 / 1h: 0

30,222

5m: 0 / 1h: 30,222

1,847

5m: 1,847 / 1h: 0

30,223

5m: 0 / 1h: 30,223

30,223

5m: 0 / 1h: 30,223

缓存读取 0 0 16,627 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 16:33:39 ~ 16:33:44 (4.0s) 16:31:52 ~ 16:31:57 (3.3s) 16:32:11 ~ 16:32:26 (10.2s) 16:32:18 ~ 16:32:28 (5.8s) 16:31:48 ~ 16:31:53 (1.8s)

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) 2,431 ms 2,005 ms 7,326 ms 2,840 ms -

首文本(SSE delta) 2,432 ms 2,007 ms 7,683 ms 2,841 ms -

总耗时 2,904 ms 2,299 ms 7,683 ms 3,342 ms -

输入 2 2 2,071 2 -

输出 28 28 23 28 -

缓存创建

1,643

5m: 1,643 / 1h: 0

1,643

5m: 0 / 1h: 1,643

1,863

5m: 1,863 / 1h: 0

1,643

5m: 0 / 1h: 1,643

缓存读取 30,222 30,222 16,772 30,223 -

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

失败

评测失败

步骤 3: cutoff text

时间 16:36:27 ~ 16:36:37 (5.3s) 16:35:07 ~ 16:35:12 (3.3s) 16:36:00 ~ 16:36:10 (7.8s) 16:35:50 ~ 16:35:55 (3.9s) -

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,826 ms 2,027 ms 5,433 ms 2,088 ms -

首文本(SSE delta) 3,828 ms 2,029 ms 5,434 ms 2,091 ms -

总耗时 4,239 ms 2,254 ms 5,670 ms 2,717 ms -

输入 2 2 2,077 2 -

输出 6 6 4 6 -

缓存创建

5m: 54 / 1h: 0

5m: 0 / 1h: 71

1,868

5m: 1,868 / 1h: 0

5m: 0 / 1h: 71

缓存读取 31,865 31,865 16,821 31,866 -

响应

2026-01

2025-01

2026-01

2026-01

未执行

步骤 4: identity_free text

时间 16:37:45 ~ 16:37:50 (4.2s) 16:38:29 ~ 16:38:34 (3.9s) 16:40:07 ~ 16:40:17 (5.8s) 16:39:34 ~ 16:39:39 (4.3s) -

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) 2,126 ms 1,911 ms 4,492 ms 1,714 ms -

首文本(SSE delta) 2,128 ms 1,913 ms 4,493 ms 1,715 ms -

总耗时 3,289 ms 2,896 ms 4,556 ms 3,189 ms -

输入 2 2 2,081 2 -

输出 69 61 28 62 -

缓存创建

5m: 39 / 1h: 0

5m: 0 / 1h: 35

1,872

5m: 1,872 / 1h: 0

5m: 0 / 1h: 39

缓存读取 31,919 31,936 16,856 31,937 -

输出速度 59.4 tok/s 62.1 tok/s - 42.1 tok/s -

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我可以协助编程、写作、分析问题和解答疑问，致力于提供有帮助、诚实且安全的回应。

我是 Claude，由 Anthropic 公司打造的 AI 助手。当前模型为 Claude Opus 4.8，可协助完成软件工程等各类任务。

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

我是 Claude，由 Anthropic 公司开发的 AI 助手。我可以帮助你处理软件工程任务、回答问题、分析代码和协助各类工作。

未执行

步骤 5: knowledge_recall text

时间 16:41:22 ~ 16:41:32 (9.7s) 16:40:38 ~ 16:40:48 (6.6s) 16:43:21 ~ 16:43:41 (16.5s) 16:42:11 ~ 16:42:21 (8.8s) -

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) 3,570 ms 2,490 ms 10,183 ms 2,387 ms -

首文本(SSE delta) 6,393 ms 3,441 ms 10,184 ms 5,428 ms -

总耗时 8,885 ms 5,646 ms 15,301 ms 7,820 ms -

输入 2 2 335 2 -

输出 689 421 387 683 -

缓存创建

4,781

5m: 4,781 / 1h: 0

4,786

5m: 0 / 1h: 4,786

301

5m: 301 / 1h: 0

4,792

5m: 0 / 1h: 4,792

缓存读取 0 0 2,714 0 -

输出速度 burst burst 75.6 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

未执行

总计

总输入(含缓存)

132,207

132,242

86,158

132,256

31,677

仅前 2 步

总输出

796

520

443

783

仅前 2 步

总缓存创建

36,739

5m: 36,739 / 1h: 0

36,757

5m: 0 / 1h: 36,757

7,751

5m: 7,751 / 1h: 0

36,768

5m: 0 / 1h: 36,768

30,223

5m: 0 / 1h: 30,223

仅前 2 步

总缓存读取

94,006

94,023

69,790

94,026

仅前 2 步

输出速度

59.4 tok/s

62.1 tok/s

75.6 tok/s

42.1 tok/s

仅前 2 步

总执行时间

30.5s

20.5s

46.4s

26.4s

7.8s

仅前 2 步

总墙钟时间

10m 59.8s

含步骤间等待

10m 11.8s

含步骤间等待

13m 6.8s

含步骤间等待

11m 50.8s

含步骤间等待

1m 25.2s

含步骤间等待

仅前 2 步

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2177

Anthropic: Claude Opus 4.8

$0.2971

Anthropic: Claude Opus 4.8

$0.1193

Anthropic: Claude Opus 4.8

$0.3037

Anthropic: Claude Opus 4.8

$0.1963

Anthropic: Claude Opus 4.8 (部分)

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall ping

平台 direct (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 - cloudflare - -

ID 格式 msg_01DgADDHJcGn5U81nmUd7w5z msg_01TTgyVas19iM36ZFBXZcQEW msg_01B9PP1h1h2R5hgwFp1iFZPKV4 msg_01RHLyQHR5DYKUMQ31qi4hcQ msg_01RvkKw9kfJHoxUEmB9WQ7kv

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
cc	cf-ray: a0c0467f3c99734b-NRT server: cloudflare set-cookie: _cfuvid=O_W7ExycwMJT4C.hbUzuwld6oPAjoPOlXfwYTtghOUA-1781512883.538366-1.0.1.1-XYTgLFiFNV2Kb6fmQREzQ_teD4gdmq6VcHHv0iVLgDw; 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	hdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011Cc4kHHDzD7MjPXZY1VUeh a0c0467f3c99734b-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_011Cc4kE2V8qtKKpqDmmU5E5
M-Kiro/Antigravity	cf-ray: a0c04969d8460adc-AMS server: cloudflare x-oneapi-request-id: 202606150843228973267618268d9d6lc1KCnDW	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	a0c04969d8460adc-AMS
O-Max	server: nginx/1.24.0 (Ubuntu) x-oneapi-request-id: 20260615164212516381282sm7GURVa	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 x-oneapi-request-id: 202606150830309436250098268d9d6l14Oaesq	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 91 96 79 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 0.0 /10 ×15 10.0 /10 ×15 0.0 /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 0.0 /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 0.0 /10 ×5 10.0 /10 ×5 —

延迟基线 10.0 /10 ×5 10.0 /10 ×5 9.0 /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 91 v3.16.0 基准（参考） 79 v3.16.0 97 v3.16.0 执行失败，不评分

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 5116 ms 3085 ms 7593 ms 3851 ms 3674 ms

输出速度 59.4 tok/s 62.1 tok/s 75.6 tok/s 42.1 tok/s -

完成步骤 5/5 成功 5/5 成功 5/5 成功 5/5 成功 1/5 成功 · 1 失败 · 3 跳过

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2177

$0.2971

$0.1193

$0.3037

$0.1963 (部分)

缓存读取占比 ⓘ 71% 71% 81% 71% 0% (部分)

揭晓

通道

服务商 FoxCode ↗

通道cc

官方基准

服务商 luminai ↗

通道M-Kiro/Antigravity

服务商 RunAPI ↗

通道O-Max

服务商 Novaris ↗

通道O-Max

返回列表