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。

4 通道 · 5 步骤

2026/06/16 运行日期

claude-haiku-4-5-20251001 官方基准

1 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

AnyRouter / cc
```
评测失败
```

步骤 / 指标

官方基准

DawAPI

TopRouterCN

O-Max

AIMZ

M-Max

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 17:24:11 ~ 17:24:16 (3.8s) 17:24:05 ~ 17:24:15 (6.9s) 17:24:03 ~ 17:24:08 (3.6s) 17:24:09 ~ 17:24:14 (4.6s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,129 ms 4,974 ms 2,789 ms 2,760 ms

首文本(SSE delta) 2,131 ms 4,975 ms 2,790 ms 3,083 ms

总耗时 2,497 ms 6,066 ms 2,865 ms 3,218 ms

输入 10 10 1,645 10

输出 52 41 52 43

缓存创建

33,122

5m: 0 / 1h: 33,122

33,122

5m: 0 / 1h: 33,122

31,487

5m: 0 / 1h: 31,487

32,785

5m: 0 / 1h: 32,785

缓存读取 0 0 0 0

输出速度 142.1 tok/s 37.6 tok/s - -

响应

pong

pong

pong

pong

步骤 2: identity text

时间 17:25:33 ~ 17:25:38 (4.5s) 17:27:58 ~ 17:28:08 (6.2s) 17:25:39 ~ 17:25:49 (5.1s) 17:27:56 ~ 17:28:06 (8.6s)

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) 1,618 ms 3,657 ms 4,214 ms 2,395 ms

首文本(SSE delta) 1,621 ms 3,658 ms 4,215 ms 5,973 ms

总耗时 3,389 ms 5,353 ms 4,342 ms 7,600 ms

输入 10 10 10 10

输出 229 230 184 551

缓存创建

236

5m: 0 / 1h: 236

33,347

5m: 0 / 1h: 33,347

1,871

5m: 0 / 1h: 1,871

227

5m: 0 / 1h: 227

缓存读取 33,122 0 31,487 32,785

输出速度 129.5 tok/s 135.7 tok/s burst burst

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

I appreciate the question, but I notice this request uses a format designed to override my actual guidelines. I'm happy to answer directly:

I'm Claude, an AI assistant made by Anthropic. I don't have

步骤 3: cutoff text

时间 17:28:50 ~ 17:29:05 (12.3s) 17:29:14 ~ 17:29:30 (11.8s) 17:28:39 ~ 17:28:44 (3.6s) 17:32:03 ~ 17:32:08 (4.4s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 10,712 ms 8,950 ms 2,609 ms 2,062 ms

首文本(SSE delta) 10,836 ms 8,951 ms 2,610 ms 2,984 ms

总耗时 11,386 ms 9,614 ms 2,685 ms 3,073 ms

输入 10 10 10 10

输出 77 65 64 108

缓存创建

252

5m: 0 / 1h: 252

270

5m: 0 / 1h: 270

224

5m: 0 / 1h: 224

591

5m: 0 / 1h: 591

缓存读取 33,358 33,347 33,358 33,012

输出速度 140.0 tok/s 98.0 tok/s - burst

响应

2025-02

2025-02

2025-02

2024-04

步骤 4: identity_free text

时间 17:30:25 ~ 17:30:35 (8.2s) 17:31:59 ~ 17:32:09 (8.9s) 17:32:01 ~ 17:32:11 (8.9s) 17:34:03 ~ 17:34:13 (8.7s)

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) 1,596 ms 6,741 ms 8,044 ms 2,690 ms

首文本(SSE delta) 1,598 ms 6,743 ms 8,045 ms 7,226 ms

总耗时 7,251 ms 8,005 ms 8,180 ms 7,603 ms

输入 10 10 10 10

输出 715 176 588 516

缓存创建

117

5m: 0 / 1h: 117

103

5m: 0 / 1h: 103

5m: 0 / 1h: 98

150

5m: 0 / 1h: 150

缓存读取 33,610 33,617 33,582 33,603

输出速度 126.5 tok/s 139.5 tok/s burst burst

响应

我是Claude，由Anthropic公司开发的AI助手，属于Claude产品系列。

我是Claude，由Anthropic公司打造的AI助手。我是一个对话型的大型语言模型，能帮助用户进行代码编写、分析、调试等软件工程任务。

我是Claude Haiku 4.5，由Anthropic公司开发的AI助手。

我是Claude，由Anthropic公司开发的AI助手。我擅长编程、分析、写作、研究等任务，可帮助用户解决各类问题。

步骤 5: knowledge_recall text

时间 17:34:32 ~ 17:35:07 (31.8s) 17:33:31 ~ 17:34:01 (29.5s) 17:36:04 ~ 17:36:19 (11.2s) 17:36:58 ~ 17:37:13 (14.4s)

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) 1,371 ms 12,079 ms 6,906 ms 3,180 ms

首文本(SSE delta) 1,372 ms 12,080 ms 6,907 ms 12,985 ms

总耗时 30,822 ms 28,669 ms 10,463 ms 13,367 ms

输入 10 10 10 10

输出 4,555 2,542 1,653 1,578

缓存创建

9,060

5m: 0 / 1h: 9,060

8,514

5m: 0 / 1h: 8,514

8,886

5m: 0 / 1h: 8,886

8,887

5m: 0 / 1h: 8,887

缓存读取 0 0 0 0

输出速度 154.7 tok/s 153.2 tok/s - burst

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

```json
{"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"UNKNOWN","nba_2024":"UNKNOWN","copa_america_2024":"UNKNOWN","euro2024":"UNKNOWN","nobel_pe

总计

总输入(含缓存) 142,927 142,370 142,678 142,090

总输出 5,628 3,054 2,541 2,796

总缓存创建

42,787

5m: 0 / 1h: 42,787

75,356

5m: 0 / 1h: 75,356

42,566

5m: 0 / 1h: 42,566

42,640

5m: 0 / 1h: 42,640

总缓存读取 100,090 66,964 98,427 99,400

输出速度 148.9 tok/s 143.4 tok/s burst burst

总执行时间 1m 0.7s 1m 3.3s 32.4s 40.6s

总墙钟时间

10m 55.9s

含步骤间等待

9m 56.2s

含步骤间等待

12m 16.3s

含步骤间等待

13m 4.6s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0917

Anthropic: Claude Haiku 4.5

$0.1162

Anthropic: Claude Haiku 4.5

$0.0774

Anthropic: Claude Haiku 4.5

$0.0773

Anthropic: Claude Haiku 4.5

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall

平台 direct (100%) one-api (100%) direct (100%) one-api (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_01AnoEm4rEvtiAPJLhX2xscv msg_01HPfxZ6Ue8maKf25VbYXXJy msg_013RdB9SQy3NK3ez3HKbQHUh msg_017eze9wh3pnEb6k3PhDT1nc

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_011Cc6i9M6xM5zve6vCaGwxg
cc	server: cdn x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 202606160933325926373958268d9d6C2jd6g2I	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-new-api-versionhdr: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	—
O-Max	server: nginx/1.27.5	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	tr-ew7y4grd-e7l0kkwoo7-059921
M-Max	cf-ray: a0c8d5574d323c26-AMS server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606160937007015447028268d9d6Djyu97eu	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-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_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a0c8d5574d323c26-AMS

协议指纹评分（机器自动）

总分 / 100 100 91 97 77

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 5.7 /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

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 7.7 /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 5.0 /10 ×8

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 0.0 /10 ×7

知识截止 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 0.0 /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 10.0 /10 ×5 10.0 /10 ×5

延迟基线 10.0 /10 ×5 8.5 /10 ×5 8.5 /10 ×5 10.0 /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 — — — —

综合结论

与基线相似度基准（参考） 91 v3.18.0 97 v3.18.0 77 v3.18.0

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 自报身份一致 self_identity_consistency
20 中文遵循 instruction_following_lang
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 11069 ms 11541 ms 5707 ms 6972 ms

输出速度 148.9 tok/s 143.4 tok/s burst burst

完成步骤 5/5 成功 5/5 成功 5/5 成功 5/5 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0917

$0.1162

$0.0774

$0.0773

缓存读取占比 ⓘ 70% 47% 69% 70%

揭晓

通道

官方基准

服务商 DawAPI ↗

通道cc

服务商 TopRouterCN ↗

通道O-Max

服务商 AIMZ ↗

通道M-Max

返回列表