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。

4 通道 · 6 步骤

2026/06/30 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

FastCode / cc
```
评测失败
```

步骤 / 指标

sucui

O-Max

SAIAi

O-Max

AnyAIGC

M-

官方基准

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 v2.1.195 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 18:04:52 ~ 18:05:02 (5.6s) 18:04:56 ~ 18:05:02 (4.4s) 18:04:50 ~ 18:05:00 (5.6s) 18:04:58 ~ 18:05:03 (3.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,190 ms 2,474 ms 4,503 ms 1,935 ms

首文本(SSE delta) 4,192 ms 2,475 ms 4,506 ms 1,937 ms

总耗时 4,222 ms 2,768 ms 4,506 ms 2,078 ms

输入 2,326 2,326 2,326 2,326

输出 4 4 4 4

缓存创建

1,711

5m: 0 / 1h: 1,711

1,711

5m: 0 / 1h: 1,711

5m: 0 / 1h: 0

缓存读取 12,732 12,732 14,491 14,443

响应

pong

pong

pong

pong

步骤 2: identity text

时间 18:06:21 ~ 18:06:31 (5.3s) 18:06:51 ~ 18:06:56 (2.9s) 18:08:18 ~ 18:08:28 (8.2s) 18:06:10 ~ 18:06:20 (7.1s)

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) 2,313 ms 1,531 ms 5,228 ms 5,539 ms

首文本(SSE delta) 2,393 ms 1,532 ms 5,229 ms 5,551 ms

总耗时 3,021 ms 1,903 ms 6,888 ms 6,155 ms

输入 2 2 2 2

输出 28 28 28 28

缓存创建

2,506

5m: 0 / 1h: 2,506

2,506

5m: 0 / 1h: 2,506

2,506

5m: 0 / 1h: 2,506

2,506

5m: 0 / 1h: 2,506

缓存读取 14,443 14,443 14,491 14,443

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 18:07:31 ~ 18:07:41 (5.0s) 18:08:12 ~ 18:08:17 (4.2s) 18:09:49 ~ 18:09:54 (3.8s) 18:09:47 ~ 18:09:52 (3.1s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,603 ms 2,936 ms 2,639 ms 1,846 ms

首文本(SSE delta) 2,604 ms 2,937 ms 2,640 ms 1,847 ms

总耗时 3,415 ms 3,184 ms 2,640 ms 2,140 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

5m: 0 / 1h: 54

5m: 0 / 1h: 64

5m: 0 / 1h: 62

5m: 0 / 1h: 54

缓存读取 16,949 16,949 16,997 16,949

响应

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 18:11:00 ~ 18:11:10 (8.9s) 18:09:55 ~ 18:10:00 (3.7s) 18:13:13 ~ 18:13:23 (8.5s) 18:13:11 ~ 18:13:16 (3.5s)

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) 5,899 ms 1,618 ms 5,871 ms 1,626 ms

首文本(SSE delta) 5,900 ms 1,620 ms 5,872 ms 1,628 ms

总耗时 7,860 ms 2,634 ms 7,347 ms 2,434 ms

输入 2 2 2 2

输出 77 61 62 59

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 37

缓存读取 17,003 17,013 17,059 17,003

输出速度 39.3 tok/s 60.2 tok/s 42.0 tok/s 73.2 tok/s

响应

你好！我是 Claude，由 Anthropic 开发的 AI 助手。我能帮你处理软件工程、写作、分析等各类任务，擅长理解需求并高效完成工作。很高兴为你服务！

我是 Claude，由 Anthropic 开发的 AI 助手，擅长编程、写作、分析与问答，可作为软件工程智能体协助完成各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手，基于 Opus 4.8 模型，可帮助你处理编程、写作、分析等各类任务。

我是 Claude，属于 Claude 产品系列，由 Anthropic 公司开发。具体模型是 Opus 4.8（claude-opus-4-8）。

步骤 5: knowledge_recall text

时间 18:12:28 ~ 18:12:38 (8.9s) 18:11:10 ~ 18:11:20 (8.0s) 18:15:58 ~ 18:16:18 (16.0s) 18:16:24 ~ 18:16:39 (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) 2,007 ms 1,510 ms 3,447 ms 3,478 ms

首文本(SSE delta) 3,379 ms 2,924 ms 6,628 ms 4,886 ms

总耗时 7,836 ms 7,053 ms 12,247 ms 10,055 ms

输入 2 2 2 2

输出 691 580 716 698

缓存创建

5,756

5m: 0 / 1h: 5,756

5,750

5m: 0 / 1h: 5,750

5,749

5m: 0 / 1h: 5,749

5,736

5m: 0 / 1h: 5,736

缓存读取 0 0 0 0

输出速度 burst 140.5 tok/s 127.4 tok/s 135.0 tok/s

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

步骤 6: digit_count text

时间 18:15:37 ~ 18:15:47 (5.2s) 18:14:39 ~ 18:14:49 (7.5s) 18:18:36 ~ 18:18:46 (5.9s) 18:19:14 ~ 18:19:19 (2.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=115479261280988160409223380478121887914699981658619441649389464459609391

HTTP 头(TTFB) 3,166 ms 5,938 ms 3,811 ms 1,398 ms

首文本(SSE delta) 3,167 ms 5,940 ms 3,812 ms 1,400 ms

总耗时 4,256 ms 6,489 ms 4,687 ms 1,826 ms

输入 211 211 596 211

输出 19 19 14 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

响应

{"n":71,"c":[5,3,12]}

{"n":71,"c":[5,3,12]}

{"n":72,"c":[6,3,9]}

{"n":71,"c":[5,3,11]}

总计

总输入(含缓存) 73,738 73,752 74,324 73,716

总输出 825 698 830 814

总缓存创建

10,066

5m: 0 / 1h: 10,066

10,070

5m: 0 / 1h: 10,070

8,356

5m: 0 / 1h: 8,356

8,333

5m: 0 / 1h: 8,333

总缓存读取 61,127 61,137 63,038 62,838

输出速度 39.3 tok/s 124.6 tok/s 109.7 tok/s 126.7 tok/s

总执行时间 38.9s 30.6s 48.0s 30.6s

总墙钟时间

10m 55.1s

含步骤间等待

9m 53.0s

含步骤间等待

13m 56.0s

含步骤间等待

14m 21.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1268

Anthropic: Claude Opus 4.8

$0.1237

Anthropic: Claude Opus 4.8

$0.1191

Anthropic: Claude Opus 4.8

$0.1166

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 one-api (100%) direct (100%) direct (100%) direct (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_01FDwnfUhm59M8jHJxjGcKTw msg_01V14XKAu3fYReznNifSg2zw msg_01qhJYJZ9CYmAhf5iqS631bV msg_01LB2Px78SJ8EUGx9JnsP1mS

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	via: 1.1 Caddy cf-ray: a13c69348dc9d361-FRA server: cloudflare x-new-api-version: v0.12.10 x-oneapi-request-id: 20260630101539552296208268d9d66oymvlgL	id_format: msg_prefix via_chain: ["1.1 Caddy"] 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-rayhdr:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a13c69348dc9d361-FRA
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_011CcZGXdNBw39oTsiv45o43
M-	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	—
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_011CcZGstrQhPK8urV4jGNFH

协议指纹评分（机器自动）

总分 / 100 97 100 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

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 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15

延迟基线 7.8 /10 ×15 10.0 /10 ×15 7.5 /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

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

知识截止 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

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

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

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

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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度 97 v3.26.0 100 v3.26.0 93 v3.26.0 基准（参考）

雷达图例

共 21 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 Thinking 体量 thinking_volume_match
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 档位思考量 tier_thinking_volume_match
21 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 5102 ms 4005 ms 6386 ms 4115 ms

输出速度 39.3 tok/s 124.6 tok/s 109.7 tok/s 126.7 tok/s

完成步骤 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 ⓘ

$0.1268

$0.1237

$0.1191

$0.1166

缓存读取占比 ⓘ 83% 83% 85% 85%

揭晓

通道

服务商 sucui ↗

通道O-Max

服务商 SAIAi ↗

通道O-Max

服务商 AnyAIGC ↗

通道M-

官方基准

返回列表