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。

5 通道 · 6 步骤

2026/06/22 运行日期

claude-opus-4-8 官方基准

全部 ≥ 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

ClaudeAPI

O-Bdrk

Novaris

O-Max

SudoCode

M-Max

官方基准

ByteCatCode

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 20:00:18 ~ 20:00:28 (6.7s) 20:00:20 ~ 20:00:25 (3.1s) 20:00:16 ~ 20:00:21 (4.2s) 20:00:24 ~ 20:00:29 (2.9s) 20:00:22 ~ 20:00:27 (4.9s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,104 ms 1,788 ms 2,582 ms 1,789 ms 2,799 ms

首文本(SSE delta) 4,880 ms 1,789 ms 2,584 ms 1,791 ms -

总耗时 4,881 ms 1,948 ms 2,793 ms 1,927 ms 3,134 ms

输入 1,491 1,454 1,454 1,454 1,454

输出 4 4 4 4 4

缓存创建

29,410

5m: 29,410 / 1h: 0

30,223

5m: 0 / 1h: 30,223

30,222

5m: 0 / 1h: 30,222

30,222

5m: 0 / 1h: 30,222

30,255

5m: 30,255 / 1h: 0

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 20:04:15 ~ 20:04:25 (7.1s) 20:03:29 ~ 20:03:34 (4.9s) 20:02:14 ~ 20:02:19 (3.5s) 20:02:49 ~ 20:02:54 (3.7s) 20:02:03 ~ 20:02:08 (4.8s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 4,050 ms 3,528 ms 1,955 ms 2,267 ms 2,552 ms

首文本(SSE delta) 4,281 ms 3,529 ms 1,956 ms 2,271 ms -

总耗时 4,631 ms 3,963 ms 2,249 ms 2,652 ms 3,007 ms

输入 2 2 2 2 2

输出 28 28 28 28 28

缓存创建

1,683

5m: 1,683 / 1h: 0

1,646

5m: 0 / 1h: 1,646

1,646

5m: 0 / 1h: 1,646

1,646

5m: 0 / 1h: 1,646

1,646

5m: 1,646 / 1h: 0

缓存读取 29,410 30,223 30,222 30,222 30,255

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

vendor: anthropic
brand: claude
model: claude-opus-4-8

步骤 3: cutoff text

时间 20:05:54 ~ 20:06:09 (10.7s) 20:06:06 ~ 20:06:11 (3.1s) 20:04:34 ~ 20:04:39 (3.5s) 20:04:10 ~ 20:04:15 (3.4s) 20:03:51 ~ 20:03:56 (4.7s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 5,164 ms 2,056 ms 2,190 ms 2,197 ms 3,176 ms

首文本(SSE delta) 5,165 ms 2,057 ms 2,192 ms 2,199 ms -

总耗时 6,331 ms 2,309 ms 2,419 ms 2,352 ms 3,437 ms

输入 2 2 2 2 2

输出 6 6 6 6 6

缓存创建

5m: 54 / 1h: 0

5m: 0 / 1h: 64

5m: 0 / 1h: 54

5m: 0 / 1h: 62

5m: 64 / 1h: 0

缓存读取 31,093 31,869 31,868 31,868 31,901

响应

2026-01

2026-01

2026-01

2025-01

2026-01

步骤 4: identity_free text

时间 20:07:10 ~ 20:07:20 (5.9s) 20:08:23 ~ 20:08:28 (3.6s) 20:07:00 ~ 20:07:10 (5.1s) 20:07:06 ~ 20:07:11 (4.0s) 20:06:41 ~ 20:06:51 (7.7s)

提示词

查看本步 Prompt

用中文说说你的身份：什么产品？由谁打造？50 字以内。

HTTP 头(TTFB) 3,645 ms 1,656 ms 2,709 ms 1,886 ms 4,779 ms

首文本(SSE delta) 3,647 ms 1,657 ms 2,710 ms 1,887 ms -

总耗时 4,588 ms 2,829 ms 3,754 ms 2,912 ms 6,521 ms

输入 2 2 2 2 2

输出 59 61 61 65 67

缓存创建

5m: 35 / 1h: 0

5m: 0 / 1h: 35

5m: 0 / 1h: 39

5m: 0 / 1h: 31

5m: 31 / 1h: 0

缓存读取 31,147 31,933 31,922 31,930 31,965

输出速度 62.7 tok/s 52.0 tok/s 58.4 tok/s 63.4 tok/s 38.5 tok/s

响应

我是 Claude Code，Anthropic 打造的官方命令行编程助手，底层模型为 Claude Opus 4.8，帮助完成软件工程任务。

我是 Claude，由 Anthropic 打造的 AI 助手，当前模型为 Claude Opus 4.8，可协助编程、研究、写作等多种任务。

我是 Claude，由 Anthropic 开发的 AI 助手，基于 Claude Opus 4.8 模型，可协助编程、写作、分析等多种任务。

我是 Claude，一个由 Anthropic 开发的人工智能助手。当前模型为 Claude Opus 4.8，可协助处理编程、写作、分析等多种任务。

我是 Claude，由 Anthropic 开发的人工智能助手，具体为 Claude Opus 4.8 模型。我可以协助你完成编程、写作、分析等多种任务。

步骤 5: knowledge_recall text

时间 20:10:57 ~ 20:11:12 (11.2s) 20:11:22 ~ 20:11:32 (8.9s) 20:09:05 ~ 20:09:15 (9.4s) 20:09:57 ~ 20:10:07 (5.8s) 20:07:59 ~ 20:08:14 (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) 2,955 ms 2,263 ms 2,440 ms 2,033 ms 4,036 ms

首文本(SSE delta) 6,048 ms 6,034 ms 5,768 ms 2,491 ms -

总耗时 9,018 ms 8,183 ms 8,188 ms 4,800 ms 9,926 ms

输入 2 2 2 2 2

输出 663 732 709 377 666

缓存创建

4,846

5m: 4,846 / 1h: 0

4,783

5m: 0 / 1h: 4,783

4,776

5m: 0 / 1h: 4,776

4,780

5m: 0 / 1h: 4,780

4,817

5m: 4,817 / 1h: 0

缓存读取 0 0 0 0 0

输出速度 burst burst burst burst 113.1 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

{"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 20:14:38 ~ 20:14:53 (13.2s) 20:13:03 ~ 20:13:08 (2.5s) 20:12:25 ~ 20:12:30 (2.9s) 20:13:45 ~ 20:13:55 (5.5s) 20:10:59 ~ 20:11:04 (4.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=069869958895897867733973685197253157137377814064269832892889482218330323

HTTP 头(TTFB) 1,693 ms 1,252 ms 1,436 ms 4,209 ms 2,388 ms

首文本(SSE delta) 10,630 ms 1,254 ms 1,437 ms 4,211 ms -

总耗时 11,067 ms 1,813 ms 1,768 ms 4,571 ms 3,079 ms

输入 251 212 211 211 245

输出 1,056 19 19 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0 0

输出速度 burst - - - -

响应

{"n":72,"c":[3,9,10]}

{"n":72,"c":[5,9,12]}

{"n":72,"c":[5,11,11]}

{"n":72,"c":[3,11,11]}

{"n":72,"c":[5,11,11]}

总计

总输入(含缓存) 129,428 132,450 132,422 132,434 132,641

总输出 1,816 850 827 499 790

总缓存创建

36,028

5m: 36,028 / 1h: 0

36,751

5m: 0 / 1h: 36,751

36,737

5m: 0 / 1h: 36,737

36,741

5m: 0 / 1h: 36,741

36,813

5m: 36,813 / 1h: 0

总缓存读取 91,650 94,025 94,012 94,020 94,121

输出速度 62.7 tok/s 52.0 tok/s 58.4 tok/s 63.4 tok/s 96.0 tok/s

总执行时间 54.8s 26.1s 28.5s 25.3s 37.8s

总墙钟时间

14m 35.1s

含步骤间等待

12m 48.0s

含步骤间等待

12m 14.0s

含步骤间等待

13m 31.0s

含步骤间等待

10m 41.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2407

Anthropic: Claude Opus 4.8

$0.3063

Anthropic: Claude Opus 4.8

$0.3057

Anthropic: Claude Opus 4.8

$0.2975

Anthropic: Claude Opus 4.8

$0.2191

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 digit_count digit_count digit_count digit_count digit_count

平台 one-api (100%) one-api (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%) Anthropic 直连 (100%)

CDN - - - - -

ID 格式 msg_01Dms7JLPwHxfUVXkxBu7wUM msg_01CqKJe5rS8EN9WgtSX2g7Lw msg_017A2j3y8qusVrYNPU86Cvuk msg_012f2Y3kvY6USxwEHaXjCWa1 msg_01C56HZhq45jCr6hP9AdRamc

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: 202606221214404195098728268d9d64zb190Yv	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	—
O-Max	server: nginx x-oneapi-request-id: 202606221213038472010878268d9d6ujTYGrSj	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	—
M-Max	server: nginx x-new-api-version: v0.0.0 x-oneapi-request-id: 20260622201226373371137ybbYjUIo	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	—
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_011CcJH92uhudG242hwFPxd7
O-Max	server: nginx/1.29.8 x-oneapi-request-id: 202606221211008564707738268d9d6tHdWRsvk	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 86 95 95 90 93

基线状态已对照基线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 7.8 /10 ×20 7.9 /10 ×20 7.9 /10 ×20 7.9 /10 ×20 7.9 /10 ×20

缓存 TTL 一致 7.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 7.0 /10 ×15

模型匹配 10.0 /10 ×14 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 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 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 10.0 /10 ×8

系统提示纯净 5.0 /10 ×8 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 10.0 /10 ×7

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

service_tier 10.0 /10 ×6 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 10.0 /10 ×5

延迟基线 10.0 /10 ×5 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 未计入

Req-ID 透传 0.0 /10 ×4 0.0 /10 ×4 0.0 /10 ×4 10.0 /10 ×4 0.0 /10 ×4

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

SDK 一致 10.0 /10 ×2 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 86 v3.24.0 95 v3.24.0 95 v3.24.0 基准（参考） 93 v3.24.0

雷达图例

共 22 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 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 中文遵循 instruction_following_lang
21 档位思考量 tier_thinking_volume_match
22 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 6753 ms 3508 ms 3529 ms 3202 ms 4851 ms

输出速度 62.7 tok/s 52.0 tok/s 58.4 tok/s 63.4 tok/s 96.0 tok/s

完成步骤 6/6 成功 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 ⓘ

$0.2407

$0.3063

$0.3057

$0.2975

$0.2191

缓存读取占比 ⓘ 71% 71% 71% 71% 71%

揭晓

通道

服务商 ClaudeAPI ↗

通道O-Bdrk

服务商 Novaris ↗

通道O-Max

服务商 SudoCode ↗

通道M-Max

官方基准

服务商 ByteCatCode ↗

通道O-Max

返回列表