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

2026/07/01 运行日期

claude-opus-4-8 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

ClaudeAPI

O-Bdrk

PackyCode

cc-Zeus

sucui

O-Max-NL

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 13:00:03 ~ 13:00:13 (6.2s) 13:00:05 ~ 13:00:10 (3.9s) 13:00:07 ~ 13:00:17 (6.6s) 13:00:09 ~ 13:00:19 (5.2s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,992 ms 2,747 ms 3,877 ms 3,938 ms

首文本(SSE delta) 3,994 ms 2,748 ms 3,883 ms 3,941 ms

总耗时 4,716 ms 2,892 ms 4,787 ms 4,221 ms

输入 2,326 2,326 2 2,326

输出 4 4 4 4

缓存创建

1,711

5m: 0 / 1h: 1,711

1,714

5m: 0 / 1h: 1,714

37,198

5m: 0 / 1h: 37,198

5m: 0 / 1h: 0

缓存读取 12,732 12,732 0 14,443

响应

pong

pong

pong

pong

步骤 2: identity text

时间 13:01:58 ~ 13:02:08 (7.4s) 13:04:10 ~ 13:04:15 (3.6s) 13:04:16 ~ 13:04:26 (5.4s) 13:01:24 ~ 13:01:29 (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。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) 6,150 ms 2,092 ms 3,899 ms 3,310 ms

首文本(SSE delta) 6,151 ms 2,093 ms 3,900 ms 3,311 ms

总耗时 6,151 ms 2,608 ms 4,313 ms 3,788 ms

输入 5,963 2 2 2

输出 68 28 28 28

缓存创建

11,072

5m: 11,072 / 1h: 0

2,506

5m: 0 / 1h: 2,506

182

5m: 0 / 1h: 182

2,506

5m: 0 / 1h: 2,506

缓存读取 0 14,446 37,198 14,443

输出速度 burst - - -

响应

I'm Claude, an AI assistant made by Anthropic.

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 13:04:30 ~ 13:04:40 (5.5s) 13:08:13 ~ 13:08:18 (4.0s) 13:07:52 ~ 13:08:02 (5.8s) 13:02:30 ~ 13:02:35 (3.1s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 4,029 ms 2,451 ms 4,368 ms 1,671 ms

首文本(SSE delta) 4,030 ms 2,452 ms 4,369 ms 1,673 ms

总耗时 4,031 ms 2,739 ms 4,605 ms 2,066 ms

输入 2 2 2 2

输出 6 6 6 6

缓存创建

2,684

5m: 0 / 1h: 2,684

5m: 0 / 1h: 62

5m: 0 / 1h: 64

5m: 0 / 1h: 62

缓存读取 14,443 16,952 37,380 16,949

响应

2026-01

2026-01

2025-01

2026-01

步骤 4: identity_free text

时间 13:07:05 ~ 13:07:15 (5.7s) 13:12:13 ~ 13:12:18 (4.1s) 13:10:36 ~ 13:10:46 (7.6s) 13:04:12 ~ 13:04:22 (6.1s)

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) 4,036 ms 2,044 ms 5,555 ms 3,831 ms

首文本(SSE delta) 4,037 ms 2,045 ms 5,556 ms 3,832 ms

总耗时 4,366 ms 3,161 ms 6,474 ms 5,216 ms

输入 5,994 2 2 2

输出 58 59 63 71

缓存创建

11,128

5m: 11,128 / 1h: 0

5m: 0 / 1h: 39

缓存读取 0 17,014 37,444 17,011

输出速度 - 52.9 tok/s 68.6 tok/s 51.3 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。

我是 Claude，由 Anthropic 开发的 AI 助手，擅长编程、写作与分析，致力于提供有帮助、诚实且安全的协助。

我是 Claude，由 Anthropic 开发的 AI 助手。这里以 Claude Code 命令行工具的形式为你提供编程和技术任务方面的帮助。

我是 Claude，由 Anthropic 开发的 AI 助手。我能协助编程、写作、分析和解答问题，擅长软件工程任务，致力于提供有用、可靠的帮助。

步骤 5: knowledge_recall text

时间 13:10:13 ~ 13:10:28 (10.0s) 13:14:06 ~ 13:14:16 (9.9s) 13:13:48 ~ 13:13:58 (6.1s) 13:06:46 ~ 13:07:01 (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) 2,300 ms 1,684 ms 3,015 ms 3,774 ms

首文本(SSE delta) 3,727 ms 3,172 ms 3,016 ms 5,185 ms

总耗时 8,723 ms 7,792 ms 5,154 ms 9,995 ms

输入 2 2 2 2

输出 693 655 343 656

缓存创建

5,991

5m: 0 / 1h: 5,991

5,749

5m: 0 / 1h: 5,749

38,093

5m: 0 / 1h: 38,093

5,758

5m: 0 / 1h: 5,758

缓存读取 0 0 0 0

输出速度 138.7 tok/s 141.8 tok/s burst 136.4 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 13:14:26 ~ 13:14:56 (28.4s) 13:15:45 ~ 13:15:55 (5.3s) 13:15:04 ~ 13:15:29 (22.2s) 13:10:27 ~ 13:10:32 (4.6s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=009592690246405425848796857309642708153122776636782976799557452724030433

HTTP 头(TTFB) 26,449 ms 2,794 ms 5,616 ms 3,228 ms

首文本(SSE delta) 26,450 ms 2,795 ms 6,926 ms 3,230 ms

总耗时 26,957 ms 3,356 ms 13,101 ms 3,638 ms

输入 211 211 2 211

输出 19 19 674 19

缓存创建

5m: 0 / 1h: 0

302

5m: 0 / 1h: 302

5m: 0 / 1h: 0

缓存读取 0 0 36,705 0

输出速度 - - 109.1 tok/s -

响应

{"n":72,"c":[8,7,9]}

{"n":72,"c":[7,7,10]}

{"n":72,"c":[8,10,8]}

{"n":72,"c":[5,7,8]}

总计

总输入(含缓存) 74,259 73,759 224,617 73,756

总输出 848 771 1,118 784

总缓存创建

32,586

5m: 22,200 / 1h: 10,386

10,070

5m: 0 / 1h: 10,070

75,878

5m: 0 / 1h: 75,878

8,365

5m: 0 / 1h: 8,365

总缓存读取 27,175 61,144 148,727 62,846

输出速度 138.7 tok/s 124.5 tok/s 103.9 tok/s 117.4 tok/s

总执行时间 1m 3.1s 30.7s 53.7s 34.7s

总墙钟时间

14m 53.0s

含步骤间等待

15m 49.9s

含步骤间等待

15m 22.0s

含步骤间等待

10m 23.6s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2589

Anthropic: Claude Opus 4.8

$0.1255

Anthropic: Claude Opus 4.8

$0.5766

Anthropic: Claude Opus 4.8

$0.1160

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%) 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 cloudflare -

ID 格式 msg_01P7myTbeCWBJ44XEqAPDuTr msg_01RA3ML6P4WHYud5zTgAgFKC msg_01NJ5y2rneuzbXvRmYyVszYQ msg_01456SfvCrSTqQjmPrVP5hWs

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-Bdrk	x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 202607010514275619604348268d9d6rlK5enhg	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-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
cc-Zeus	via: 1.1 Caddy cf-ray: a142ef531a7e0d77-LAX server: cloudflare x-oneapi-request-id: 2026070113154780076378bIeo7QIS	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-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	a142ef531a7e0d77-LAX
O-Max-NL	via: 1.1 Caddy cf-ray: a142ee8098d3d2f6-FRA server: cloudflare x-new-api-version: v0.12.10 x-oneapi-request-id: 202607010515134485134418268d9d6UVviuQsM	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	a142ee8098d3d2f6-FRA
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_011Ccam96Y4UaFgAntgy4gFQ

协议指纹评分（机器自动）

总分 / 100 75 99 78 100

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 0.0 /10 ×20 10.0 /10 ×20 7.3 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 4.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15

延迟基线 9.2 /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 5.3 /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) 0.0 /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 0.0 /10 ×5 10.0 /10 ×5

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

流式投递 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 0.0 /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

综合结论

与基线相似度 75 v3.26.0 99 v3.26.0 78 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 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 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 9157 ms 3758 ms 6406 ms 4821 ms

输出速度 138.7 tok/s 124.5 tok/s 103.9 tok/s 117.4 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 ⓘ

$0.2589

$0.1255

$0.5766

$0.1160

缓存读取占比 ⓘ 37% 83% 66% 85%

揭晓

通道

服务商 ClaudeAPI ↗

通道O-Bdrk

服务商 PackyCode ↗

通道cc-Zeus

服务商 sucui ↗

通道O-Max-NL

官方基准

返回列表