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

2026/06/19 运行日期

claude-haiku-4-5-20251001 官方基准

2 通道分数 < 70 另含 1 任务失败

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

DBAI / M-
```
评测失败
```

步骤 / 指标

官方基准

duckcoding

cc专用

FoxCode

R-ultra

FastCode

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code

模型 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001 claude-haiku-4-5-20251001

客户端 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 20:24:07 ~ 20:24:12 (3.5s) 20:24:01 ~ 20:24:11 (5.1s) 20:23:59 ~ 20:24:04 (4.7s) 20:24:03 ~ 20:24:18 (14.8s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,192 ms 3,193 ms 2,744 ms 12,092 ms

首文本(SSE delta) 2,194 ms 3,196 ms 2,745 ms 12,421 ms

总耗时 2,491 ms 4,233 ms 3,755 ms 13,294 ms

输入 10 10 33,022 10

输出 40 47 85 63

缓存创建

33,122

5m: 0 / 1h: 33,122

33,122

5m: 0 / 1h: 33,122

5m: 0 / 1h: 0

33,122

5m: 0 / 1h: 33,122

缓存读取 0 0 0 0

输出速度 134.7 tok/s 45.3 tok/s 84.2 tok/s 72.2 tok/s

响应

pong

pong

pong

pong

步骤 2: identity text

时间 20:26:25 ~ 20:26:35 (6.7s) 20:27:36 ~ 20:27:46 (5.3s) 20:26:46 ~ 20:26:56 (5.0s) 20:25:35 ~ 20:25:50 (11.9s)

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。请只输出以下 3 行内容（label 小写，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) 1,822 ms 2,268 ms 2,199 ms 3,185 ms

首文本(SSE delta) 1,824 ms 2,269 ms 2,369 ms 3,186 ms

总耗时 5,586 ms 4,586 ms 4,070 ms 9,547 ms

输入 10 10 33,611 10

输出 244 195 226 225

缓存创建

220

5m: 0 / 1h: 220

227

5m: 0 / 1h: 227

5m: 0 / 1h: 0

178

5m: 0 / 1h: 178

缓存读取 33,122 33,122 0 33,122

输出速度 64.9 tok/s 84.2 tok/s 132.9 tok/s 35.4 tok/s

响应

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

vendor: anthropic
brand: claude
model: claude-haiku-4-5-20251001

步骤 3: cutoff text

时间 20:28:47 ~ 20:28:52 (4.2s) 20:30:21 ~ 20:30:36 (14.2s) 20:29:49 ~ 20:29:54 (3.4s) 20:29:45 ~ 20:29:55 (8.0s)

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) 2,650 ms 10,872 ms 1,833 ms 3,786 ms

首文本(SSE delta) 2,652 ms 10,956 ms 1,834 ms 3,794 ms

总耗时 3,262 ms 12,607 ms 2,519 ms 4,426 ms

输入 10 10 34,204 10

输出 80 114 72 70

缓存创建

286

5m: 0 / 1h: 286

237

5m: 0 / 1h: 237

5m: 0 / 1h: 0

5m: 0 / 1h: 42

缓存读取 33,342 33,349 0 33,300

输出速度 131.1 tok/s 69.0 tok/s 105.1 tok/s 110.8 tok/s

响应

2025-02

2025-02

2025-02

2025-02

步骤 4: identity_free text

时间 20:32:29 ~ 20:32:39 (5.9s) 20:32:44 ~ 20:32:54 (5.8s) 20:32:10 ~ 20:32:20 (7.0s) 20:33:08 ~ 20:33:18 (7.7s)

提示词

查看本步 Prompt

请用中文介绍你自己是什么产品或助手，由谁开发，50 字以内。

HTTP 头(TTFB) 2,750 ms 2,113 ms 2,348 ms 4,764 ms

首文本(SSE delta) 2,752 ms 2,114 ms 2,349 ms 4,765 ms

总耗时 4,884 ms 4,694 ms 5,825 ms 6,537 ms

输入 10 10 34,633 10

输出 339 347 563 121

缓存创建

122

5m: 0 / 1h: 122

152

5m: 0 / 1h: 152

5m: 0 / 1h: 0

5m: 0 / 1h: 43

缓存读取 33,628 33,586 0 33,342

输出速度 159.0 tok/s 134.5 tok/s 162.0 tok/s 68.3 tok/s

响应

我是Claude，一个由Anthropic公司开发的AI助手。我可以帮助用户完成各种编程和软件工程任务，包括代码编写、调试、测试和代码审查。我的知识截止日期是2025年2月。

我是Claude Haiku 4.5，由Anthropic公司开发的AI助手。擅长软件工程、编程、分析等多种任务。

我是Claude，由Anthropic公司开发的AI助手。

我是 Claude，由 Anthropic 公司开发的 AI 助手。当前运行的是 Claude Haiku 4.5 模型，专注于帮助用户完成软件工程和各类编程任务。

步骤 5: knowledge_recall text

时间 20:35:08 ~ 20:35:53 (33.0s) 20:33:54 ~ 20:34:19 (24.8s) 20:35:50 ~ 20:36:15 (20.2s) 20:35:33 ~ 20:35:53 (19.9s)

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) 1,487 ms 1,693 ms 1,998 ms 3,266 ms

首文本(SSE delta) 1,488 ms 1,694 ms 1,999 ms 3,268 ms

总耗时 31,786 ms 23,845 ms 18,931 ms 17,452 ms

输入 10 10 10,530 10

输出 4,705 3,559 2,807 916

缓存创建

8,707

5m: 0 / 1h: 8,707

8,703

5m: 0 / 1h: 8,703

5m: 0 / 1h: 0

8,063

5m: 0 / 1h: 8,063

缓存读取 0 0 0 0

输出速度 155.3 tok/s 160.7 tok/s 165.8 tok/s 64.6 tok/s

响应

```json
{"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

```json
{"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

```json
{"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

{"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 20:36:57 ~ 20:37:23 (22.2s) 20:35:21 ~ 20:35:46 (24.6s) 20:38:22 ~ 20:38:37 (11.4s) 20:38:48 ~ 20:39:38 (45.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=010432176914101220083961471190905674767778815055574522024178757350116607

HTTP 头(TTFB) 963 ms 1,623 ms 1,365 ms 19,342 ms

首文本(SSE delta) 978 ms 1,625 ms 1,366 ms 19,344 ms

总耗时 21,181 ms 23,284 ms 10,614 ms 43,990 ms

输入 193 193 209 193

输出 4,014 4,480 2,305 2,324

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 198.7 tok/s 206.8 tok/s 249.2 tok/s 94.3 tok/s

响应

```json
{"n":72,"c":[11,12,4]}
```

```json
{"n":72,"c":[11,12,4]}
```

```json
{"n":72,"c":[11,12,4]}
```

```json
{"n":72,"c":[11,12,4]}
```

总计

总输入(含缓存) 142,792 142,741 146,209 141,455

总输出 9,422 8,742 6,058 3,719

总缓存创建

42,457

5m: 0 / 1h: 42,457

42,441

5m: 0 / 1h: 42,441

5m: 0 / 1h: 0

41,448

5m: 0 / 1h: 41,448

总缓存读取 100,092 100,057 0 99,764

输出速度 164.4 tok/s 170.1 tok/s 183.3 tok/s 76.7 tok/s

总执行时间 1m 15.5s 1m 19.8s 51.6s 1m 47.4s

总墙钟时间

13m 15.1s

含步骤间等待

11m 45.1s

含步骤间等待

14m 38.1s

含步骤间等待

15m 34.3s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1104

Anthropic: Claude Haiku 4.5

$0.1070

Anthropic: Claude Haiku 4.5

$0.1765

Anthropic: Claude Haiku 4.5

$0.0806

Anthropic: Claude Haiku 4.5

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

平台 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%)

CDN - - cloudflare -

ID 格式 msg_01ChopDHwbozBvc5UJusZ6et msg_01CBDXNJU2kRXQmFkGo9UQg1 msg_01YPxn2aTajh6sr81GfvkssT msg_vrtx_01RKgG1MQaX54pDufn22miLs

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
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: true usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011CcCdV7nWZQJeUrMRL9pu8
cc专用	server: Nginx x-oneapi-request-id: 202606191235235152041158268d9d6J2tSX7Ly	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: true usage_has_cache_fields: true	hdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
R-ultra	cf-ray: a0e2972dab9efd4f-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606191238238442584018268d9d6mXPzifVS	id_format: msg_prefix has_service_tier: true has_inference_geo: true model_has_date_suffix: true 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:model:date_suffixbody:usage.cache_fieldsupstream:anthropic-direct-from-body	a0e2972dab9efd4f-NRT
cc	server: nginx	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: false has_inference_geo: false model_has_date_suffix: true usage_has_cache_fields: true	body:id:msg_prefixbody:model:date_suffixbody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—

协议指纹评分（机器自动）

总分 / 100 100 98 69 83

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 0.0 /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

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 5.0 /10 ×14 10.0 /10 ×14

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

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

知识截止 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 0.0 /10 ×6

inference_geo 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 0.0 /10 ×5

延迟基线 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 7.4 /10 ×5

Req-ID 透传 10.0 /10 ×4 0.0 /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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度基准（参考） 98 v3.24.0 69 v3.24.0 83 v3.24.0

雷达图例

共 23 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 自报身份一致 self_identity_consistency
20 信封自报一致 envelope_self_report_match
21 中文遵循 instruction_following_lang
22 档位思考量 tier_thinking_volume_match
23 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 11532 ms 12208 ms 7619 ms 15874 ms

输出速度 164.4 tok/s 170.1 tok/s 183.3 tok/s 76.7 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 ⓘ

$0.1104

$0.1070

$0.1765

$0.0806

缓存读取占比 ⓘ 70% 70% 0% 71%

揭晓

通道

官方基准

服务商 duckcoding ↗

通道cc专用

服务商 FoxCode ↗

通道R-ultra

服务商 FastCode ↗

通道cc

返回列表