Linux 性能测试工具 sysbench 的安装与简单使用

本文涉及的产品
RDS MySQL Serverless 基础系列,0.5-2RCU 50GB
云数据库 RDS MySQL,集群系列 2核4GB
推荐场景:
搭建个人博客
云原生数据库 PolarDB PostgreSQL 版,标准版 2核4GB 50GB
简介: Linux 性能测试工具 sysbench 的安装与简单使用一 背景sysbench是一款开源的多线程性能测试工具,可以执行CPU/内存/线程/IO/数据库等方面的性能测试。sysbench 支持以下几种测试模式 :1、CPU运算性能2、内存分配及传输速度3、磁盘IO性能4、POSIX线程性能5、互斥性测试6、数据库性能(OLTP基准测试)。

Linux 性能测试工具 sysbench 的安装与简单使用

一 背景

sysbench是一款开源的多线程性能测试工具,可以执行CPU/内存/线程/IO/数据库等方面的性能测试。
sysbench 支持以下几种测试模式 :

  • 1、CPU运算性能
  • 2、内存分配及传输速度
  • 3、磁盘IO性能
  • 4、POSIX线程性能
  • 5、互斥性测试
  • 6、数据库性能(OLTP基准测试)。目前sysbench主要支持 MySQL,PostgreSQL 等几种数据库。

二 实验环境

2.1 操作系统

[gysl@gysl-DevOps ~]$ cat /etc/centos-release
CentOS Linux release 7.6.1810 (Core) 

2.2 其他配置

安装EPEL,安装方法之前的文章有介绍。

三 安装

安装命令如下:

[gysl@gysl-DevOps ~]$ sudo yum -y install sysbench

四 简单使用过程

4.1 查看软件版本

[gysl@gysl-DevOps ~]$ sysbench --version
sysbench 1.0.9

4.2 查看系统帮助

[gysl@gysl-DevOps ~]$ sysbench --help

See 'sysbench --test=<name> help' for a list of options for each test. #查看每个测试项目的更多选项列表
Usage:
sysbench [options]... [testname] [command]

Commands implemented by most tests: prepare run cleanup help

General options:
  --threads=N                     number of threads to use [1] #创建测试线程的数目。默认为1。  
  --events=N                      limit for total number of events [0]
  --time=N                        limit for total execution time in seconds [10]
  --forced-shutdown=STRING        number of seconds to wait after the --time limit before forcing shutdown, or 'off' to disable [off] #超过max-time强制中断。默认是off。  
  --thread-stack-size=SIZE        size of stack per thread [64K] #每个线程的堆栈大小。默认是32K。  
  --rate=N                        average transactions rate. 0 for unlimited rate [0]
  --report-interval=N             periodically report intermediate statistics with a specified interval in seconds. 0 disables intermediate reports [0]
  --report-checkpoints=[LIST,...] dump full statistics and reset all counters at specified points in time. The argument is a list of comma-separated values representing the amount of time in seconds elapsed from start of test when report checkpoint(s) must be performed. Report checkpoints are off by default. []
  --debug[=on|off]                print more debugging info [off]
  --validate[=on|off]             perform validation checks where possible [off]
  --help[=on|off]                 print help and exit [off]
  --version[=on|off]              print version and exit [off]
  --config-file=FILENAME          File containing command line options
  --tx-rate=N                     deprecated alias for --rate [0]
  --max-requests=N                deprecated alias for --events [0] #请求的最大数目。默认为10000,0代表不限制。  
  --max-time=N                    deprecated alias for --time [0] #最大执行时间,单位是s。默认是0,不限制。  
  --num-threads=N                 deprecated alias for --threads [1]

Pseudo-Random Numbers Generator options:
  --rand-type=STRING random numbers distribution {uniform,gaussian,special,pareto} [special]
  --rand-spec-iter=N number of iterations used for numbers generation [12]
  --rand-spec-pct=N  percentage of values to be treated as 'special' (for special distribution) [1]
  --rand-spec-res=N  percentage of 'special' values to use (for special distribution) [75]
  --rand-seed=N      seed for random number generator. When 0, the current time is used as a RNG seed. [0]
  --rand-pareto-h=N  parameter h for pareto distribution [0.2]

Log options:
  --verbosity=N verbosity level {5 - debug, 0 - only critical messages} [3]

  --percentile=N       percentile to calculate in latency statistics (1-100). Use the special value of 0 to disable percentile calculations [95]
  --histogram[=on|off] print latency histogram in report [off]

General database options:

  --db-driver=STRING  specifies database driver to use ('help' to get list of available drivers)
  --db-ps-mode=STRING prepared statements usage mode {auto, disable} [auto]
  --db-debug[=on|off] print database-specific debug information [off]

Compiled-in database drivers:
  mysql - MySQL driver
  pgsql - PostgreSQL driver

mysql options:
  --mysql-host=[LIST,...]          MySQL server host [localhost]
  --mysql-port=[LIST,...]          MySQL server port [3306]
  --mysql-socket=[LIST,...]        MySQL socket
  --mysql-user=STRING              MySQL user [sbtest]
  --mysql-password=STRING          MySQL password []
  --mysql-db=STRING                MySQL database name [sbtest]
  --mysql-ssl[=on|off]             use SSL connections, if available in the client library [off]
  --mysql-ssl-cipher=STRING        use specific cipher for SSL connections []
  --mysql-compression[=on|off]     use compression, if available in the client library [off]
  --mysql-debug[=on|off]           trace all client library calls [off]
  --mysql-ignore-errors=[LIST,...] list of errors to ignore, or "all" [1213,1020,1205]
  --mysql-dry-run[=on|off]         Dry run, pretend that all MySQL client API calls are successful without executing them [off]

pgsql options:
  --pgsql-host=STRING     PostgreSQL server host [localhost]
  --pgsql-port=N          PostgreSQL server port [5432]
  --pgsql-user=STRING     PostgreSQL user [sbtest]
  --pgsql-password=STRING PostgreSQL password []
  --pgsql-db=STRING       PostgreSQL database name [sbtest]

Compiled-in tests:
  fileio - File I/O test
  cpu - CPU performance test
  memory - Memory functions speed test
  threads - Threads subsystem performance test
  mutex - Mutex performance test

See 'sysbench <testname> help' for a list of options for each test.

4.3 测试过程

sysbench 的测试过程一般分为三个阶段:

  • prepare:准备阶段,准备测试数据。
  • run:执行测试阶段。
  • cleanup:清理垃圾数据阶段。

4.4 CPU 性能测试

找出指定范围内最大质数,时间越短 CPU 性能越好。

4.4.1 查看帮助信息

[gysl@gysl-DevOps ~]$ sysbench --test=cpu help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

cpu options:
  --cpu-max-prime=N upper limit for primes generator [10000]

4.4.2 测试过程

[gysl@gysl-DevOps ~]$ sudo sysbench --test=cpu --cpu-max-prime=5000 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Prime numbers limit: 5000

Initializing worker threads...

Threads started!

CPU speed:
    events per second:  3059.82

General statistics:
    total time:                          10.0002s
    total number of events:              30603

Latency (ms):
         min:                                  0.29
         avg:                                  0.33
         max:                                  6.10
         95th percentile:                      0.50
         sum:                               9979.54

Threads fairness:
    events (avg/stddev):           30603.0000/0.00
    execution time (avg/stddev):   9.9795/0.00

本次测试中,线程数为1,质数个数为5000。

4.5 内存测试

4.5.1 查看帮助信息

[gysl@gysl-DevOps ~]$ sysbench --test=memory help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

memory options:
  --memory-block-size=SIZE    size of memory block for test [1K]
  --memory-total-size=SIZE    total size of data to transfer [100G]
  --memory-scope=STRING       memory access scope {global,local} [global]
  --memory-hugetlb[=on|off]   allocate memory from HugeTLB pool [off]
  --memory-oper=STRING        type of memory operations {read, write, none} [write]
  --memory-access-mode=STRING memory access mode {seq,rnd} [seq]

4.5.2 测试过程

[gysl@gysl-DevOps ~]$ free -h
              total        used        free      shared  buff/cache   available
Mem:           972M        137M        313M        7.6M        521M        637M
Swap:            0B          0B          0B
[gysl@gysl-DevOps ~]$ sudo sysbench --test=memory --memory-block-size=8k --memory-total-size=972M run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Running memory speed test with the following options:
  block size: 8KiB
  total size: 972MiB
  operation: write
  scope: global

Initializing worker threads...

Threads started!

Total operations: 124416 (1603394.35 per second)

972.00 MiB transferred (12526.52 MiB/sec)

General statistics:
    total time:                          0.0761s
    total number of events:              124416

Latency (ms):
         min:                                  0.00
         avg:                                  0.00
         max:                                  0.93
         95th percentile:                      0.00
         sum:                                 61.24

Threads fairness:
    events (avg/stddev):           124416.0000/0.00
    execution time (avg/stddev):   0.0612/0.00

4.6 磁盘I/O测试

4.6.1 查看帮助信息

[gysl@gysl-DevOps ~]$ sysbench --test=fileio help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

fileio options:
  --file-num=N              number of files to create [128]
  --file-block-size=N       block size to use in all IO operations [16384]
  --file-total-size=SIZE    total size of files to create [2G]
  --file-test-mode=STRING   test mode {seqwr, seqrewr, seqrd, rndrd, rndwr, rndrw}
  --file-io-mode=STRING     file operations mode {sync,async,mmap} [sync]
  --file-async-backlog=N    number of asynchronous operatons to queue per thread [128]
  --file-extra-flags=STRING additional flags to use on opening files {sync,dsync,direct} []
  --file-fsync-freq=N       do fsync() after this number of requests (0 - don't use fsync()) [100]
  --file-fsync-all[=on|off] do fsync() after each write operation [off]
  --file-fsync-end[=on|off] do fsync() at the end of test [on]
  --file-fsync-mode=STRING  which method to use for synchronization {fsync, fdatasync} [fsync]
  --file-merged-requests=N  merge at most this number of IO requests if possible (0 - don't merge) [0]
  --file-rw-ratio=N         reads/writes ratio for combined test [1.5]

4.6.2 测试过程

[gysl@gysl-DevOps ~]$ sysbench --test=fileio help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

fileio options:
  --file-num=N              number of files to create [128]
  --file-block-size=N       block size to use in all IO operations [16384]
  --file-total-size=SIZE    total size of files to create [2G]
  --file-test-mode=STRING   test mode {seqwr, seqrewr, seqrd, rndrd, rndwr, rndrw}
  --file-io-mode=STRING     file operations mode {sync,async,mmap} [sync]
  --file-async-backlog=N    number of asynchronous operatons to queue per thread [128]
  --file-extra-flags=STRING additional flags to use on opening files {sync,dsync,direct} []
  --file-fsync-freq=N       do fsync() after this number of requests (0 - don't use fsync()) [100]
  --file-fsync-all[=on|off] do fsync() after each write operation [off]
  --file-fsync-end[=on|off] do fsync() at the end of test [on]
  --file-fsync-mode=STRING  which method to use for synchronization {fsync, fdatasync} [fsync]
  --file-merged-requests=N  merge at most this number of IO requests if possible (0 - don't merge) [0]
  --file-rw-ratio=N         reads/writes ratio for combined test [1.5]

[gysl@gysl-DevOps ~]$ sudo sysbench --test=fileio --num-threads=1 --file-total-size=1G --file-test-mode=rndrw prepare
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

128 files, 8192Kb each, 1024Mb total
Creating files for the test...
Extra file open flags: 0
Creating file test_file.0
Creating file test_file.1
.......................
[gysl@gysl-DevOps ~]$ sudo sysbench --test=fileio --num-threads=1 --file-total-size=1G --file-test-mode=rndrw run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Extra file open flags: 0
128 files, 8MiB each
1GiB total file size
Block size 16KiB
Number of IO requests: 0
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Initializing worker threads...

Threads started!

File operations:
    reads/s:                      2042.32
    writes/s:                     1361.55
    fsyncs/s:                     4351.23

Throughput:
    read, MiB/s:                  31.91
    written, MiB/s:               21.27

General statistics:
    total time:                          10.0190s
    total number of events:              78021

Latency (ms):
         min:                                  0.00
         avg:                                  0.13
         max:                                105.19
         95th percentile:                      0.16
         sum:                               9932.94

Threads fairness:
    events (avg/stddev):           78021.0000/0.00
    execution time (avg/stddev):   9.9329/0.00
[gysl@gysl-DevOps ~]$ sudo sysbench --test=fileio --num-threads=1 --file-total-size=1G --file-test-mode=rndrw cleanup
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Removing test files...

4.7 线程测试

4.7.1 帮助信息

[gysl@gysl-DevOps ~]$ sysbench --test=threads help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

threads options:
  --thread-yields=N number of yields to do per request [1000]
  --thread-locks=N  number of locks per thread [8]

4.7.2 测试过程

[gysl@gysl-DevOps ~]$ sudo sysbench  --test=threads --num-threads=500 --thread-yields=100 --thread-locks=4 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --num-threads is deprecated, use --threads instead
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Running the test with following options:
Number of threads: 500
Initializing random number generator from current time

Initializing worker threads...

Threads started!

General statistics:
    total time:                          10.5847s
    total number of events:              50140

Latency (ms):
         min:                                  0.15
         avg:                                104.54
         max:                               1874.74
         95th percentile:                    427.07
         sum:                            5241634.75

Threads fairness:
    events (avg/stddev):           100.2800/25.55
    execution time (avg/stddev):   10.4833/0.07

4.8 Mutex 测试

4.8.1 帮助信息

[gysl@gysl-DevOps ~]$ sudo sysbench --test=mutex help
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

mutex options:
  --mutex-num=N   total size of mutex array [4096]
  --mutex-locks=N number of mutex locks to do per thread [50000]
  --mutex-loops=N number of empty loops to do outside mutex lock [10000]

4.8.2 测试过程

[gysl@gysl-DevOps ~]$ sudo sysbench --test=mutex --mutex-num=2048 --mutex-locks=20000 --mutex-loops=5000 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
sysbench 1.0.9 (using system LuaJIT 2.0.4)

Running the test with following options:
Number of threads: 1
Initializing random number generator from current time

Initializing worker threads...

Threads started!

General statistics:
    total time:                          0.0342s
    total number of events:              1

Latency (ms):
         min:                                 33.58
         avg:                                 33.58
         max:                                 33.58
         95th percentile:                     33.72
         sum:                                 33.58

Threads fairness:
    events (avg/stddev):           1.0000/0.00
    execution time (avg/stddev):   0.0336/0.00

五 总结

5.1 对于数据库的测试,本文未进行介绍,后续遇到该情况时会进行介绍。

5.2 由于时间仓促,文章中大部分英文本人并未进行翻译,但是涉及到的英文都不算难。

5.3 磁盘I/O测试中,--file-extra-flags 选项比较重要,有的存储设备是直接I/O,其他详情还请查阅相关资料。

5.4 本文仅仅根据帮助信息整理完成,其他详情还请参考官方手册。

六 参考资料

6.1 相关资料

6.2 相关手册

6.3 IO测试参考

相关实践学习
通过性能测试PTS对云服务器ECS进行规格选择与性能压测
本文为您介绍如何利用性能测试PTS对云服务器ECS进行规格选择与性能压测。
相关文章
|
2月前
|
Java 测试技术 数据安全/隐私保护
软件测试中的自动化策略与工具应用
在软件开发的快速迭代中,自动化测试以其高效、稳定的特点成为了质量保证的重要手段。本文将深入探讨自动化测试的核心概念、常见工具的应用,以及如何设计有效的自动化测试策略,旨在为读者提供一套完整的自动化测试解决方案,帮助团队提升测试效率和软件质量。
|
1月前
|
运维 关系型数据库 MySQL
os-copilot安装_配置_功能测试全集
我是一位中级运维工程师,我平时工作会涉及到 各类服务器的 数据库 与 java环境配置 操作。 我顺利使用了OS Copilot的 -t -f | 功能,我的疑惑是不能在自动操作过程中直接给与脚本运行权限,必须需要自己运行一下 chmod 这个既然有了最高的权限,为什么就不能直接给与运行权限呢。 我认为 -t 功能有用,能解决后台运行基础命令操作。 我认为 -f 功能有用,可以通过task文件中撰写连续任务操作。 我认为 | 对文件理解上有很直接的解读,可以在理解新程序上有很大帮助。
175 86
|
5天前
|
存储 人工智能 编译器
【03】鸿蒙实战应用开发-华为鸿蒙纯血操作系统Harmony OS NEXT-测试hello word效果-虚拟华为手机真机环境调试-为DevEco Studio编译器安装中文插件-测试写一个滑动块效果-介绍诸如ohos.ui等依赖库-全过程实战项目分享-从零开发到上线-优雅草卓伊凡
【03】鸿蒙实战应用开发-华为鸿蒙纯血操作系统Harmony OS NEXT-测试hello word效果-虚拟华为手机真机环境调试-为DevEco Studio编译器安装中文插件-测试写一个滑动块效果-介绍诸如ohos.ui等依赖库-全过程实战项目分享-从零开发到上线-优雅草卓伊凡
27 10
【03】鸿蒙实战应用开发-华为鸿蒙纯血操作系统Harmony OS NEXT-测试hello word效果-虚拟华为手机真机环境调试-为DevEco Studio编译器安装中文插件-测试写一个滑动块效果-介绍诸如ohos.ui等依赖库-全过程实战项目分享-从零开发到上线-优雅草卓伊凡
|
11天前
|
人工智能 Linux iOS开发
Burp Suite Professional 2025.2 (macOS, Linux, Windows) - Web 应用安全、测试和扫描
Burp Suite Professional 2025.2 (macOS, Linux, Windows) - Web 应用安全、测试和扫描
39 12
Burp Suite Professional 2025.2 (macOS, Linux, Windows) - Web 应用安全、测试和扫描
|
2天前
|
Linux 网络安全 iOS开发
Metasploit Framework 6.4.49 (macOS, Linux, Windows) - 开源渗透测试框架
Metasploit Framework 6.4.49 (macOS, Linux, Windows) - 开源渗透测试框架
12 0
Metasploit Framework 6.4.49 (macOS, Linux, Windows) - 开源渗透测试框架
|
1月前
|
弹性计算 Ubuntu Java
OS-Copilot-ubuntu镜像版本的具体测试使用(安装方式有单独注明)
作为一名个人开发者,我主要负责云资源的运维和管理。在使用OS Copilot的过程中,我遇到了一些配置问题,特别是在ECS实例中设置AccessKey时,但最终成功解决了。通过使用OS Copilot的-t/-f/管道功能,我大大提升了效率,减少了命令编写的工作量,特别是在搭建Java运行环境时效果显著。此外,| 功能帮助我快速理解文档,整体体验非常流畅,推荐给其他开发者使用。
43 6
|
1月前
|
弹性计算 运维 Ubuntu
os-copilot在Alibaba Cloud Linux镜像下的安装与功能测试
我顺利使用了OS Copilot的 -t -f 功能,我的疑惑是在换行的时候就直接进行提问了,每次只能写一个问题,没法连续换行更有逻辑的输入问题。 我认为 -t 管道 功能有用 ,能解决环境问题的连续性操作。 我认为 -f 管道 功能有用 ,可以单独创建可连续性提问的task问题。 我认为 | 对文件直接理解在新的服务器理解有很大的帮助。 此外,我还有建议 可以在非 co 的环境下也能进行连续性的提问。
79 7
|
1月前
|
安全 Linux 测试技术
Intel Linux 内核测试套件-LKVS介绍 | 龙蜥大讲堂104期
《Intel Linux内核测试套件-LKVS介绍》(龙蜥大讲堂104期)主要介绍了LKVS的定义、使用方法、测试范围、典型案例及其优势。LKVS是轻量级、低耦合且高代码覆盖率的测试工具,涵盖20多个硬件和内核属性,已开源并集成到多个社区CICD系统中。课程详细讲解了如何使用LKVS进行CPU、电源管理和安全特性(如TDX、CET)的测试,并展示了其在实际应用中的价值。
|
2月前
|
Web App开发 IDE 测试技术
Selenium:强大的 Web 自动化测试工具
Selenium 是一款强大的 Web 自动化测试工具,包括 Selenium IDE、WebDriver 和 Grid 三大组件,支持多种编程语言和跨平台操作。它能有效提高测试效率,解决跨浏览器兼容性问题,进行性能测试和数据驱动测试,尽管存在学习曲线较陡、不稳定等缺点,但其优势明显,是自动化测试领域的首选工具。
282 17
Selenium:强大的 Web 自动化测试工具
|
2月前
|
Linux Shell 网络安全
Kali Linux系统Metasploit框架利用 HTA 文件进行渗透测试实验
本指南介绍如何利用 HTA 文件和 Metasploit 框架进行渗透测试。通过创建反向 shell、生成 HTA 文件、设置 HTTP 服务器和发送文件,最终实现对目标系统的控制。适用于教育目的,需合法授权。
97 9
Kali Linux系统Metasploit框架利用 HTA 文件进行渗透测试实验

热门文章

最新文章

  • 1
    小鱼深度评测 | 通义灵码2.0,不仅可跨语言编码,自动生成单元测试,更炸裂的是集成DeepSeek模型且免费使用,太炸裂了。
  • 2
    3天功能开发→3小时:通义灵码2.0+DEEPSEEK实测报告,单元测试生成准确率92%的秘密
  • 3
    Potpie.ai:比Copilot更狠!这个AI直接接管项目代码,自动Debug+测试+开发全搞定
  • 4
    基于FPGA的图像双线性插值算法verilog实现,包括tb测试文件和MATLAB辅助验证
  • 5
    【01】噩梦终结flutter配安卓android鸿蒙harmonyOS 以及next调试环境配鸿蒙和ios真机调试环境-flutter项目安卓环境配置-gradle-agp-ndkVersion模拟器运行真机测试环境-本地环境搭建-如何快速搭建android本地运行环境-优雅草卓伊凡-很多人在这步就被难倒了
  • 6
    大前端之前端开发接口测试工具postman的使用方法-简单get接口请求测试的使用方法-简单教学一看就会-以实际例子来说明-优雅草卓伊凡
  • 7
    「ximagine」业余爱好者的非专业显示器测试流程规范,同时也是本账号输出内容的数据来源!如何测试显示器?荒岛整理总结出多种测试方法和注意事项,以及粗浅的原理解析!
  • 8
    用户说 | 通义灵码2.0,跨语言编码+自动生成单元测试+集成DeepSeek模型且免费使用
  • 9
    以项目登录接口为例-大前端之开发postman请求接口带token的请求测试-前端开发必学之一-如果要学会联调接口而不是纯写静态前端页面-这个是必学-本文以优雅草蜻蜓Q系统API为实践来演示我们如何带token请求接口-优雅草卓伊凡
  • 10
    AxBench:斯坦福大学推出评估语言模型控制方法的基准测试框架