blocks|key|917817|text|令人惊讶的是，重新启动解决了这个问题(我以为我已经尝试过了)。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|917818|注释中的解决方案罗伯特·克罗夫拉也可能对其他人有用，因为它与我第一次解决问题时所做的非常相似。|offset|length|917819|entityMap|0|LINK|mutability|MUTABLE|url|https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch#comment73133147_43022843^0|0|8|8|0|0^^$0|@$1|2|3|4|5|6|7|N|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|O|8|@]|9|@$D|P|E|Q|1|R]]|A|$]]|$1|F|3|-4|5|6|7|S|8|@]|9|@]|A|$]]]|G|$H|$5|I|J|K|A|$L|M]]]]

Surprise surprise, rebooting solved the issue (I thought I had already tried that).
The solution <a href="https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch#comment73133147_43022843">Robert Crovella mentioned</a> in the comments may also be useful to someone else, since it's pretty similar to what I did to solve the issue the first time I had it.

blocks|key|598773|text|作为埃塔尔，重新启动可以解决这个问题，但是我认为不重新启动的过程会有帮助。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|598774|关于中文，请访问我的博客->+中文版|598775|错误信息|598776|NVML:驱动程序/库版本不匹配|blockquote|598777|告诉我们Nvidia驱动程序内核模块(+kmod+)有错误的版本，所以我们应该卸载该驱动程序，然后加载正确的kmod版本。|598778|我们怎么能做到呢？|598779|首先，我们应该知道哪些驱动程序是加载的。|598780|lsmod+%7C+grep+nvidia|code-block|syntax|javascript|598781|你可能会得到|598782|nvidia_uvm++++++++++++634880++8
nvidia_drm+++++++++++++53248++0
nvidia_modeset++++++++790528++1+nvidia_drm
nvidia++++++++++++++12312576++86+nvidia_modeset,nvidia_uvm|598783|我们的最终目标是卸载nvidia模块，因此我们应该依靠nvidia卸载模块。|style|CODE|598784|sudo+rmmod+nvidia_drm
sudo+rmmod+nvidia_modeset
sudo+rmmod+nvidia_uvm|598785|然后，卸载nvidia|598786|sudo+rmmod+nvidia|598787|故障排除|598788|如果您得到像rmmod:+ERROR:+Module+nvidia+is+in+use这样的错误(这表明内核模块正在使用)，您应该终止使用kmod的进程：|598789|sudo+lsof+/dev/nvidia*|598790|然后消灭这些过程，然后继续卸载这些kmod。|598791|测试|598792|确认你成功卸载了那些kmod|598793|lsmod+%7C+grep+nvidia|598794|你什么都不应该得到。然后确认您可以加载正确的驱动程序：|598795|nvidia-smi|598796|你应该得到正确的输出。|598797|entityMap|0|LINK|mutability|MUTABLE|url|https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/43023000#43023000|1|https://comzyh.com/blog/archives/967/^0|2|3|0|0|F|3|1|0|0|0|0|0|0|0|0|0|A|6|R|6|0|0|5|6|0|0|0|6|11|0|0|0|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|23|8|@]|9|@$A|24|B|25|1|26]]|C|$]]|$1|D|3|E|5|6|7|27|8|@]|9|@$A|28|B|29|1|2A]]|C|$]]|$1|F|3|G|5|6|7|2B|8|@]|9|@]|C|$]]|$1|H|3|I|5|J|7|2C|8|@]|9|@]|C|$]]|$1|K|3|L|5|6|7|2D|8|@]|9|@]|C|$]]|$1|M|3|N|5|6|7|2E|8|@]|9|@]|C|$]]|$1|O|3|P|5|6|7|2F|8|@]|9|@]|C|$]]|$1|Q|3|R|5|S|7|2G|8|@]|9|@]|C|$T|U]]|$1|V|3|W|5|6|7|2H|8|@]|9|@]|C|$]]|$1|X|3|Y|5|S|7|2I|8|@]|9|@]|C|$T|U]]|$1|Z|3|10|5|6|7|2J|8|@$A|2K|B|2L|11|12]|$A|2M|B|2N|11|12]]|9|@]|C|$]]|$1|13|3|14|5|S|7|2O|8|@]|9|@]|C|$T|U]]|$1|15|3|16|5|6|7|2P|8|@$A|2Q|B|2R|11|12]]|9|@]|C|$]]|$1|17|3|18|5|S|7|2S|8|@]|9|@]|C|$T|U]]|$1|19|3|1A|5|6|7|2T|8|@]|9|@]|C|$]]|$1|1B|3|1C|5|6|7|2U|8|@$A|2V|B|2W|11|12]]|9|@]|C|$]]|$1|1D|3|1E|5|S|7|2X|8|@]|9|@]|C|$T|U]]|$1|1F|3|1G|5|6|7|2Y|8|@]|9|@]|C|$]]|$1|1H|3|1I|5|6|7|2Z|8|@]|9|@]|C|$]]|$1|1J|3|1K|5|6|7|30|8|@]|9|@]|C|$]]|$1|1L|3|1M|5|S|7|31|8|@]|9|@]|C|$T|U]]|$1|1N|3|1O|5|6|7|32|8|@]|9|@]|C|$]]|$1|1P|3|1Q|5|S|7|33|8|@]|9|@]|C|$T|U]]|$1|1R|3|1S|5|6|7|34|8|@]|9|@]|C|$]]|$1|1T|3|-4|5|6|7|35|8|@]|9|@]|C|$]]]|1U|$1V|$5|1W|1X|1Y|C|$1Z|20]]|21|$5|1W|1X|1Y|C|$1Z|22]]]]

As <a href="https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/43023000#43023000">etal said</a>, rebooting can solve this problem, but I think a procedure without rebooting will help.
For Chinese, check my blog -&gt; <a href="https://comzyh.com/blog/archives/967/" rel="noreferrer">中文版</a>
The error message
<blockquote>
NVML: Driver/library version mismatch
</blockquote>
tell us the Nvidia driver kernel module (kmod) have a wrong version, so we should unload this driver, and then load the correct version of kmod
<h2>How can we do that?</h2>
First, we should know which drivers are loaded.
<pre class="lang-none prettyprint-override"><code>lsmod | grep nvidia
</code></pre>
You may get
<pre class="lang-none prettyprint-override"><code>nvidia_uvm 634880 8
nvidia_drm 53248 0
nvidia_modeset 790528 1 nvidia_drm
nvidia 12312576 86 nvidia_modeset,nvidia_uvm
</code></pre>
Our final goal is to unload <code>nvidia</code> mod, so we should unload the module depend on <code>nvidia</code>:
<pre class="lang-none prettyprint-override"><code>sudo rmmod nvidia_drm
sudo rmmod nvidia_modeset
sudo rmmod nvidia_uvm
</code></pre>
Then, unload <code>nvidia</code>
<pre class="lang-none prettyprint-override"><code>sudo rmmod nvidia
</code></pre>
<h2>Troubleshooting</h2>
If you get an error like <code>rmmod: ERROR: Module nvidia is in use</code>, which indicates that the kernel module is in use, you should kill the process that using the kmod:
<pre class="lang-none prettyprint-override"><code>sudo lsof /dev/nvidia*
</code></pre>
and then kill those process, then continue to unload the kmods.
<h2>Test</h2>
Confirm you successfully unload those kmods
<pre class="lang-none prettyprint-override"><code>lsmod | grep nvidia
</code></pre>
You should get nothing. Then confirm you can load the correct driver:
<pre class="lang-none prettyprint-override"><code>nvidia-smi
</code></pre>
You should get the correct output.

blocks|key|917868|text|这也发生在Ubuntu16.04上使用nvidia-348包(Ubuntu16.04上最新的Nvidia版本)上。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|917869|但是，我可以通过通过nvidia-390安装专有GPU驱动程序PPA来解决这个问题。|917870|因此，Ubuntu+16.04上描述的问题的解决方案是这样做的：|917871|sudo+add-apt-repository+ppa:graphics-drivers/ppa|unordered-list-item|917872|sudo+apt-get+update|917873|sudo+apt-get+install+nvidia-390|917874|注意：本指南假定Ubuntu安装干净。如果以前安装了驱动程序，则可能需要重新启动才能重新加载所有内核模块。|BOLD|917875|entityMap|0|LINK|mutability|MUTABLE|url|https://launchpad.net/%257Egraphics-drivers/%2Barchive/ubuntu/ppa?field.series_filter=xenial^0|J|A|0|A|A|M|C|0|0|0|0|1C|0|0|J|0|0|V|0|0|3|0^^$0|@$1|2|3|4|5|6|7|11|8|@$9|12|A|13|B|C]]|D|@]|E|$]]|$1|F|3|G|5|6|7|14|8|@$9|15|A|16|B|C]]|D|@$9|17|A|18|1|19]]|E|$]]|$1|H|3|I|5|6|7|1A|8|@]|D|@]|E|$]]|$1|J|3|K|5|L|7|1B|8|@$9|1C|A|1D|B|C]]|D|@]|E|$]]|$1|M|3|N|5|L|7|1E|8|@$9|1F|A|1G|B|C]]|D|@]|E|$]]|$1|O|3|P|5|L|7|1H|8|@$9|1I|A|1J|B|C]]|D|@]|E|$]]|$1|Q|3|R|5|6|7|1K|8|@$9|1L|A|1M|B|S]]|D|@]|E|$]]|$1|T|3|-4|5|6|7|1N|8|@]|D|@]|E|$]]]|U|$V|$5|W|X|Y|E|$Z|10]]]]

This also happened to me on Ubuntu 16.04 using the <code>nvidia-348</code> package (latest Nvidia version on Ubuntu 16.04).
However I could resolve the problem by installing <code>nvidia-390</code> through the <a href="https://launchpad.net/%7Egraphics-drivers/+archive/ubuntu/ppa?field.series_filter=xenial" rel="nofollow noreferrer">Proprietary GPU Drivers PPA</a>.
So a solution to the described problem on Ubuntu 16.04 is doing this:
<ul>
<li><code>sudo add-apt-repository ppa:graphics-drivers/ppa</code></li>
<li><code>sudo apt-get update</code></li>
<li><code>sudo apt-get install nvidia-390</code></li>
</ul>
Note: This guide assumes a clean Ubuntu install. If you have previous drivers installed a reboot might be needed to reload all the kernel modules.

blocks|key|919862|text|我有这个问题，其他的治疗方法都没有用。错误消息是不透明的，但是检查https://en.wikipedia.org/wiki/Dmesg的输出是关键：|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|919863|[+++10.118255]+NVRM:+API+mismatch:+the+client+has+the+version+410.79,+but
+++++++++++NVRM:+this+kernel+module+has+the+version+384.130.++Please
+++++++++++NVRM:+make+sure+that+this+kernel+module+and+all+NVIDIA+driver
+++++++++++NVRM:+components+have+the+same+version.|code-block|syntax|javascript|919864|但是，我已经完全删除了384版本，并删除了所有剩余的内核驱动程序nvidia-384*。但即使在重启之后，我还是得到了这个。看到这一点意味着内核仍然被编译为引用384，但它只找到410。所以我重新编译了我的内核：|style|CODE|919865|uname+-a+#+Find+the+kernel+it's+using

Linux+blah+4.13.0-43-generic+#48~16.04.1-Ubuntu+SMP+Thu+May+17+12:56:46+UTC+2018+x86_64+x86_64+x86_64+GNU/Linux


update-initramfs+-c+-k+4.13.0-43-generic+#+Recompile+it
reboot|919866|然后起作用了。|919867|删除384之后，我仍然有384个文件在:+/var/lib/dkms/nvidia-XXX/XXX.YY/4.13.0-43-generic/x86_64/module+/lib/modules/4.13.0-43-泛型/内核/驱动程序中。|919868|我建议使用locate命令(默认情况下不安装)，而不是每次搜索文件系统。|919869|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Dmesg^0|X|Z|0|0|0|W|B|0|0|0|0|5|6|0^^$0|@$1|2|3|4|5|6|7|12|8|@]|9|@$A|13|B|14|1|15]]|C|$]]|$1|D|3|E|5|F|7|16|8|@]|9|@]|C|$G|H]]|$1|I|3|J|5|6|7|17|8|@$A|18|B|19|K|L]]|9|@]|C|$]]|$1|M|3|N|5|F|7|1A|8|@]|9|@]|C|$G|H]]|$1|O|3|P|5|6|7|1B|8|@]|9|@]|C|$]]|$1|Q|3|R|5|6|7|1C|8|@]|9|@]|C|$]]|$1|S|3|T|5|6|7|1D|8|@$A|1E|B|1F|K|L]]|9|@]|C|$]]|$1|U|3|-4|5|6|7|1G|8|@]|9|@]|C|$]]]|V|$W|$5|X|Y|Z|C|$10|11]]]]

I was having this problem, and none of the other remedies worked. The error message was opaque, but checking the output of <a href="https://en.wikipedia.org/wiki/Dmesg" rel="noreferrer">dmesg</a> was the key:
<pre class="lang-none prettyprint-override"><code>[ 10.118255] NVRM: API mismatch: the client has the version 410.79, but
 NVRM: this kernel module has the version 384.130. Please
 NVRM: make sure that this kernel module and all NVIDIA driver
 NVRM: components have the same version.
</code></pre>
However, I had completely removed the 384 version, and removed any remaining kernel drivers <code>nvidia-384*</code>. But even after reboot, I was still getting this. Seeing this meant that the kernel was still compiled to reference 384, but it was only finding 410. So I recompiled my kernel:
<pre class="lang-none prettyprint-override"><code>uname -a # Find the kernel it's using

Linux blah 4.13.0-43-generic #48~16.04.1-Ubuntu SMP Thu May 17 12:56:46 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux


update-initramfs -c -k 4.13.0-43-generic # Recompile it
reboot
</code></pre>
And then it worked.
After removing 384, I still had 384 files in:
/var/lib/dkms/nvidia-XXX/XXX.YY/4.13.0-43-generic/x86_64/module
/lib/modules/4.13.0-43-generic/kernel/drivers
I recommend using the <code>locate</code> command (not installed by default) rather than searching the filesystem every time.

blocks|key|598916|text|我也有这个问题(我正在运行Ubuntu+18.04+(仿生海狸))。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|598917|我所做的：|598918|dpkg+-l+%7C+grep+-i+nvidia|style|CODE|598919|然后是sudo+apt-get+remove+--purge+nvidia-381+(以及每一个复制版本，在我的例子中，我有381、384和387)|598920|然后sudo+ubuntu-drivers+devices列出可用的内容。|598921|我选择了sudo+apt+install+nvidia-driver-430。|598922|在此之后，nvidia-smi提供了正确的输出(不需要重新启动)。但我想你可以在有疑问的时候重新启动。|598923|我还跟踪这个装置重新安装了cuda%2Bcudnn。|598924|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_18.04_LTS_.28Bionic_Beaver.29|1|https://gist.github.com/Mahedi-61/2a2f1579d4271717d421065168ce6a73#file-cuda_installation_on_ubuntu_18-04-L60^0|D|C|0|0|0|0|O|0|3|12|0|2|R|0|4|Y|0|5|A|0|4|4|1|0^^$0|@$1|2|3|4|5|6|7|13|8|@]|9|@$A|14|B|15|1|16]]|C|$]]|$1|D|3|E|5|6|7|17|8|@]|9|@]|C|$]]|$1|F|3|G|5|6|7|18|8|@$A|19|B|1A|H|I]]|9|@]|C|$]]|$1|J|3|K|5|6|7|1B|8|@$A|1C|B|1D|H|I]]|9|@]|C|$]]|$1|L|3|M|5|6|7|1E|8|@$A|1F|B|1G|H|I]]|9|@]|C|$]]|$1|N|3|O|5|6|7|1H|8|@$A|1I|B|1J|H|I]]|9|@]|C|$]]|$1|P|3|Q|5|6|7|1K|8|@$A|1L|B|1M|H|I]]|9|@]|C|$]]|$1|R|3|S|5|6|7|1N|8|@]|9|@$A|1O|B|1P|1|1Q]]|C|$]]|$1|T|3|-4|5|6|7|1R|8|@]|9|@]|C|$]]]|U|$V|$5|W|X|Y|C|$Z|10]]|11|$5|W|X|Y|C|$Z|12]]]]

I had the issue too (I'm running <a href="https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_18.04_LTS_.28Bionic_Beaver.29" rel="nofollow noreferrer">Ubuntu 18.04</a> (Bionic Beaver)).
What I did:
<code>dpkg -l | grep -i nvidia</code>
Then
<code>sudo apt-get remove --purge nvidia-381</code> (and every duplicate version, in my case I had 381, 384 and 387)
Then <code>sudo ubuntu-drivers devices</code> to list what's available.
And I choose <code>sudo apt install nvidia-driver-430</code>.
After that, <code>nvidia-smi</code> gave the correct output (no need to reboot). But I suppose you can reboot when in doubt.
I also followed <a href="https://gist.github.com/Mahedi-61/2a2f1579d4271717d421065168ce6a73#file-cuda_installation_on_ubuntu_18-04-L60" rel="nofollow noreferrer">this installation</a> to reinstall cuda+cudnn.

blocks|key|917927|text|前两名的答案不能解决我的问题。我在英伟达官方论坛上找到了解决方案，解决了我的问题。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|917928|下面的错误信息可能是通过不同的方法安装两个不同版本的驱动程序造成的。例如，通过APT和官方安装程序安装Nvidia驱动程序。|offset|length|917929|未能初始化NVML:驱动程序/库版本不匹配|blockquote|917930|要解决这个问题，只需要执行以下两个命令中的一个。|917931|sudo+apt-get+--purge+remove+"*nvidia*"|code-block|syntax|javascript|917932|sudo+/usr/bin/nvidia-uninstall|917933|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/APT_(software)^0|0|13|3|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|Z|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|10|8|@]|9|@$D|11|E|12|1|13]]|A|$]]|$1|F|3|G|5|H|7|14|8|@]|9|@]|A|$]]|$1|I|3|J|5|6|7|15|8|@]|9|@]|A|$]]|$1|K|3|L|5|M|7|16|8|@]|9|@]|A|$N|O]]|$1|P|3|Q|5|M|7|17|8|@]|9|@]|A|$N|O]]|$1|R|3|-4|5|6|7|18|8|@]|9|@]|A|$]]]|S|$T|$5|U|V|W|A|$X|Y]]]]

The top-2 answers can't solve my problem. I found a solution at the Nvidia official forum solved my problem.
The below error information may be caused by installing two different versions of the driver by different approaches. For example, install Nvidia driver by <a href="https://en.wikipedia.org/wiki/APT_(software)" rel="noreferrer">APT</a> and the official installer.
<blockquote>
Failed to initialize NVML: Driver/library version mismatch
</blockquote>
To solve this problem, there is only a need to execute one of the following two commands.
<pre class="lang-sh prettyprint-override"><code>sudo apt-get --purge remove &quot;*nvidia*&quot;
</code></pre>
<pre class="lang-sh prettyprint-override"><code>sudo /usr/bin/nvidia-uninstall
</code></pre>

blocks|key|598991|text|我重新安装了Nvidia驱动程序:在root模式下运行以下命令：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|598992|systemctl+isolate+multi-user.target|ordered-list-item|offset|length|style|CODE|598993|modprobe+-r+nvidia-drm|598994|重新安装Nvidia驱动程序：chmod+%2Bx+NVIDIA-Linux-x86_64–410.57.run|598995|systemctl+start+graphical.target|598996|最后检查nvidia-smi|598997|这得益于：|598998|https://medium.com/better-programming/how-to-install-nvidia-drivers-and-cuda-10-0-for-rtx-2080-ti-gpu-on-ubuntu-16-04-18-04-ce32e4edf1c0|unordered-list-item|598999|https://unix.stackexchange.com/questions/440840/how-to-unload-kernel-module-nvidia-drm|599000|entityMap|0|LINK|mutability|MUTABLE|url|1^0|0|0|Z|0|0|M|0|F|13|0|0|W|0|4|A|0|0|0|3S|0|0|0|2E|1|0^^$0|@$1|2|3|4|5|6|7|15|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|16|8|@$E|17|F|18|G|H]]|9|@]|A|$]]|$1|I|3|J|5|D|7|19|8|@$E|1A|F|1B|G|H]]|9|@]|A|$]]|$1|K|3|L|5|D|7|1C|8|@$E|1D|F|1E|G|H]]|9|@]|A|$]]|$1|M|3|N|5|D|7|1F|8|@$E|1G|F|1H|G|H]]|9|@]|A|$]]|$1|O|3|P|5|6|7|1I|8|@$E|1J|F|1K|G|H]]|9|@]|A|$]]|$1|Q|3|R|5|6|7|1L|8|@]|9|@]|A|$]]|$1|S|3|T|5|U|7|1M|8|@]|9|@$E|1N|F|1O|1|1P]]|A|$]]|$1|V|3|W|5|U|7|1Q|8|@]|9|@$E|1R|F|1S|1|1T]]|A|$]]|$1|X|3|-4|5|6|7|1U|8|@]|9|@]|A|$]]]|Y|$Z|$5|10|11|12|A|$13|T]]|14|$5|10|11|12|A|$13|W]]]]

I had reinstalled the Nvidia driver: run these commands in root mode:
<ol>
<li><code>systemctl isolate multi-user.target</code>
</li>
<li><code>modprobe -r nvidia-drm</code>
</li>
<li>Reinstall the Nvidia driver: <code>chmod +x NVIDIA-Linux-x86_64–410.57.run</code>
</li>
<li><code>systemctl start graphical.target</code>
</li>
</ol>
And finally check <code>nvidia-smi</code>
Thanks to:
<ul>
<li><a href="https://medium.com/better-programming/how-to-install-nvidia-drivers-and-cuda-10-0-for-rtx-2080-ti-gpu-on-ubuntu-16-04-18-04-ce32e4edf1c0" rel="nofollow noreferrer">How To Install Nvidia Drivers and CUDA-10.0 for RTX 2080 Ti GPU on Ubuntu-16.04/18.04</a>
</li>
<li><a href="https://unix.stackexchange.com/questions/440840/how-to-unload-kernel-module-nvidia-drm">How to unload kernel module 'nvidia-drm'?</a>
</li>
</ul>

blocks|key|599065|text|在CentOS机器上进行了正常的内核更新之后，我就遇到了这个问题。由于所有CUDA和Nvidia驱动程序和库都是通过百胜存储库安装的，所以我通过以下步骤设法解决了这些问题：|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|599066|sudo+yum+remove+nvidia-driver-*
sudo+reboot
sudo+yum+install+nvidia-driver-cuda+nvidia-modprobe
sudo+modprobe+nvidia+#+Or+just+reboot|code-block|syntax|javascript|599067|它确保了我的内核和我的Nvidia驱动程序是一致的。我认为，仅仅重新启动可能会导致加载错误版本的内核模块。|599068|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/CentOS|1|https://en.wikipedia.org/wiki/Yum_(software)^0|1|6|0|1M|2|1|0|0|0^^$0|@$1|2|3|4|5|6|7|U|8|@]|9|@$A|V|B|W|1|X]|$A|Y|B|Z|1|10]]|C|$]]|$1|D|3|E|5|F|7|11|8|@]|9|@]|C|$G|H]]|$1|I|3|J|5|6|7|12|8|@]|9|@]|C|$]]|$1|K|3|-4|5|6|7|13|8|@]|9|@]|C|$]]]|L|$M|$5|N|O|P|C|$Q|R]]|S|$5|N|O|P|C|$Q|T]]]]

I experienced this problem after a normal kernel update on a <a href="https://en.wikipedia.org/wiki/CentOS" rel="nofollow noreferrer">CentOS</a> machine. Since all CUDA and Nvidia drivers and libraries have been installed via <a href="https://en.wikipedia.org/wiki/Yum_(software)" rel="nofollow noreferrer">YUM</a> repositories, I managed to solve the issues using the following steps:
<pre class="lang-none prettyprint-override"><code>sudo yum remove nvidia-driver-*
sudo reboot
sudo yum install nvidia-driver-cuda nvidia-modprobe
sudo modprobe nvidia # Or just reboot
</code></pre>
It made sure my kernel and my Nvidia driver were consistent. I reckon that just rebooting may result in the wrong version of the kernel module being loaded.

blocks|key|917966|text|我将容器提交到Docker映像中。然后我用这个Docker映像重新创建了另一个容器，问题就解决了。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|917967|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Docker_%2528software%2529^0|7|6|0|0^^$0|@$1|2|3|4|5|6|7|L|8|@]|9|@$A|M|B|N|1|O]]|C|$]]|$1|D|3|-4|5|6|7|P|8|@]|9|@]|C|$]]]|E|$F|$5|G|H|I|C|$J|K]]]]

I committed the container into a <a href="https://en.wikipedia.org/wiki/Docker_%28software%29" rel="nofollow noreferrer">Docker</a> image. Then I recreated another container using this Docker image and the problem was gone.

blocks|key|599117|text|这些答案对我没有用：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|599118|伊塔尔的回答|unordered-list-item|offset|length|599119|Comzyh的回答|599120|用户9170的答案|599121|dmesg|style|CODE|599122|NVRM:+API+mismatch:+the+client+has+the+version+418.67,+but
NVRM:+this+kernel+module+has+the+version+430.26.++Please
NVRM:+make+sure+that+this+kernel+module+and+all+NVIDIA+driver
NVRM:+components+have+the+same+version.|code-block|syntax|javascript|599123|卸载旧驱动程序418.67并安装新驱动程序430.26+(下载NVIDIA-Linux-x86_64-430.26.run)：|599124|sudo+apt-get+--purge+remove+"*nvidia*"
sudo+/usr/bin/nvidia-uninstall
chmod+%2Bx+NVIDIA-Linux-x86_64-430.26.run
sudo+./NVIDIA-Linux-x86_64-430.26.run
[ignore+abort]|599125|cat+/proc/driver/nvidia/version|599126|NVRM+version:+NVIDIA+UNIX+x86_64+Kernel+Module++430.26++Tue+Jun++4+17:40:52+CDT+2019
GCC+version:++gcc+version+7.4.0+(Ubuntu+7.4.0-1ubuntu1~18.04.1)|599127|entityMap|0|LINK|mutability|MUTABLE|url|https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/43023000#43023000|1|https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/45319156#45319156|2|https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/54349675#54349675^0|0|0|6|0|0|0|9|1|0|0|9|2|0|0|5|0|0|7|6|L|6|V|U|0|0|0|V|0|0^^$0|@$1|2|3|4|5|6|7|1D|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|1E|8|@]|9|@$E|1F|F|1G|1|1H]]|A|$]]|$1|G|3|H|5|D|7|1I|8|@]|9|@$E|1J|F|1K|1|1L]]|A|$]]|$1|I|3|J|5|D|7|1M|8|@]|9|@$E|1N|F|1O|1|1P]]|A|$]]|$1|K|3|L|5|6|7|1Q|8|@$E|1R|F|1S|M|N]]|9|@]|A|$]]|$1|O|3|P|5|Q|7|1T|8|@]|9|@]|A|$R|S]]|$1|T|3|U|5|6|7|1U|8|@$E|1V|F|1W|M|N]|$E|1X|F|1Y|M|N]|$E|1Z|F|20|M|N]]|9|@]|A|$]]|$1|V|3|W|5|Q|7|21|8|@]|9|@]|A|$R|S]]|$1|X|3|Y|5|6|7|22|8|@$E|23|F|24|M|N]]|9|@]|A|$]]|$1|Z|3|10|5|Q|7|25|8|@]|9|@]|A|$R|S]]|$1|11|3|-4|5|6|7|26|8|@]|9|@]|A|$]]]|12|$13|$5|14|15|16|A|$17|18]]|19|$5|14|15|16|A|$17|1A]]|1B|$5|14|15|16|A|$17|1C]]]]

These answers did not work for me:
<ul>
<li><a href="https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/43023000#43023000">etal's answer</a>
</li>
<li><a href="https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/45319156#45319156">Comzyh's answer</a>
</li>
<li><a href="https://stackoverflow.com/questions/43022843/nvidia-nvml-driver-library-version-mismatch/54349675#54349675">user9170's answer</a>
</li>
</ul>
<code>dmesg</code>
<pre class="lang-none prettyprint-override"><code>NVRM: API mismatch: the client has the version 418.67, but
NVRM: this kernel module has the version 430.26. Please
NVRM: make sure that this kernel module and all NVIDIA driver
NVRM: components have the same version.
</code></pre>
Uninstall old driver <code>418.67</code> and install new driver <code>430.26</code> (download <code>NVIDIA-Linux-x86_64-430.26.run</code>):
<pre class="lang-none prettyprint-override"><code>sudo apt-get --purge remove &quot;*nvidia*&quot;
sudo /usr/bin/nvidia-uninstall
chmod +x NVIDIA-Linux-x86_64-430.26.run
sudo ./NVIDIA-Linux-x86_64-430.26.run
[ignore abort]
</code></pre>
<code>cat /proc/driver/nvidia/version</code>
<pre class="lang-none prettyprint-override"><code>NVRM version: NVIDIA UNIX x86_64 Kernel Module 430.26 Tue Jun 4 17:40:52 CDT 2019
GCC version: gcc version 7.4.0 (Ubuntu 7.4.0-1ubuntu1~18.04.1)
</code></pre>

blocks|key|2313757|text|重新启动。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2313758|如果问题仍然存在：|2313759|sudo+rmmod+nvidia_drm
sudo+rmmod+nvidia_modeset
sudo+rmmod+nvidia
nvidia-smi|code-block|syntax|javascript|2313760|CentOS和红帽企业Linux+(RHEL)：|offset|length|2313761|cd+/boot
mv+initramfs-$(uname+-r).img+/boot/initramfs-$(uname+-r).img.bak
dracut+-vf+initramfs-$(uname+-r).img+$(uname+-r)|2313762|然后|2313763|reboot|2313764|适用于Debian/Ubuntu|2313765|update-initramfs+-u|2313766|2313767|apt+install+-y+dkms+&&+dkms+install+-m+nvidia+-v+440.82|2313768|将440.82更改为实际版本。|2313769|提示:获取Nvidia驱动程序版本：|2313770|ls+/usr/src|2313771|您将找到Nvidia驱动程序目录，如nvidia-440.82。|style|CODE|2313772|此外，您还可以删除所有Nvidia包并重新安装驱动程序：|2313773|apt+purge+nvidia*
apt+purge+*cuda*

#+Check
apt+list+-i+%7Cgrep+nvidia
apt+list+-i+%7Cgrep+cuda|2313774|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/CentOS|1|https://en.wikipedia.org/wiki/Red_Hat_Enterprise_Linux|2|https://en.wikipedia.org/wiki/Debian|3|https://en.wikipedia.org/wiki/Ubuntu_%2528operating_system%2529^0|0|0|0|0|6|0|7|9|1|0|0|0|0|3|6|2|A|6|3|0|0|0|0|0|0|0|I|D|0|0|0^^$0|@$1|2|3|4|5|6|7|1R|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|1S|8|@]|9|@]|A|$]]|$1|D|3|E|5|F|7|1T|8|@]|9|@]|A|$G|H]]|$1|I|3|J|5|6|7|1U|8|@]|9|@$K|1V|L|1W|1|1X]|$K|1Y|L|1Z|1|20]]|A|$]]|$1|M|3|N|5|F|7|21|8|@]|9|@]|A|$G|H]]|$1|O|3|P|5|6|7|22|8|@]|9|@]|A|$]]|$1|Q|3|R|5|F|7|23|8|@]|9|@]|A|$G|H]]|$1|S|3|T|5|6|7|24|8|@]|9|@$K|25|L|26|1|27]|$K|28|L|29|1|2A]]|A|$]]|$1|U|3|V|5|F|7|2B|8|@]|9|@]|A|$G|H]]|$1|W|3|C|5|6|7|2C|8|@]|9|@]|A|$]]|$1|X|3|Y|5|F|7|2D|8|@]|9|@]|A|$G|H]]|$1|Z|3|10|5|6|7|2E|8|@]|9|@]|A|$]]|$1|11|3|12|5|6|7|2F|8|@]|9|@]|A|$]]|$1|13|3|14|5|F|7|2G|8|@]|9|@]|A|$G|H]]|$1|15|3|16|5|6|7|2H|8|@$K|2I|L|2J|17|18]]|9|@]|A|$]]|$1|19|3|1A|5|6|7|2K|8|@]|9|@]|A|$]]|$1|1B|3|1C|5|F|7|2L|8|@]|9|@]|A|$G|H]]|$1|1D|3|-4|5|6|7|2M|8|@]|9|@]|A|$]]]|1E|$1F|$5|1G|1H|1I|A|$1J|1K]]|1L|$5|1G|1H|1I|A|$1J|1M]]|1N|$5|1G|1H|1I|A|$1J|1O]]|1P|$5|1G|1H|1I|A|$1J|1Q]]]]

Reboot.
If the problem still exist:
<pre><code>sudo rmmod nvidia_drm
sudo rmmod nvidia_modeset
sudo rmmod nvidia
nvidia-smi
</code></pre>
For <a href="https://en.wikipedia.org/wiki/CentOS" rel="noreferrer">CentOS</a> and <a href="https://en.wikipedia.org/wiki/Red_Hat_Enterprise_Linux" rel="noreferrer">Red Hat Enterprise Linux</a> (RHEL):
<pre><code>cd /boot
mv initramfs-$(uname -r).img /boot/initramfs-$(uname -r).img.bak
dracut -vf initramfs-$(uname -r).img $(uname -r)
</code></pre>
Then
<pre><code>reboot
</code></pre>
For <a href="https://en.wikipedia.org/wiki/Debian" rel="noreferrer">Debian</a>/<a href="https://en.wikipedia.org/wiki/Ubuntu_%28operating_system%29" rel="noreferrer">Ubuntu</a>:
<pre class="lang-sh prettyprint-override"><code>update-initramfs -u
</code></pre>
If the problem persists:
<pre class="lang-sh prettyprint-override"><code>apt install -y dkms &amp;&amp; dkms install -m nvidia -v 440.82
</code></pre>
Change 440.82 to your actual version.
Tip: get the Nvidia driver version:
<pre class="lang-sh prettyprint-override"><code>ls /usr/src
</code></pre>
You will find the Nvidia driver directory, such as <code>nvidia-440.82</code>.
<hr />
Also, you can remove all Nvidia packages and reinstall the driver again:
<pre class="lang-sh prettyprint-override"><code>apt purge nvidia*
apt purge *cuda*

# Check
apt list -i |grep nvidia
apt list -i |grep cuda
</code></pre>

blocks|key|599239|text|我必须重新启动内核，并删除以前安装的所有包(在第一次安装期间)。请确保删除所有包，即使在通过以下命令删除包之后：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|599240|sudo+apt-get+--purge+remove+"*nvidia*"|code-block|syntax|javascript|599241|包，如“libtinfo6:i+386”，不会被删除。|599242|我用的是Ubuntu+20.04+(焦距Fossa)和Nvidia+440。为此，您必须删除以下图像中显示的所有包。|offset|length|599243|需要删除的所有包的列表：|599244|​|599245|📷|atomic|599246|599247|如图所示，确保您正在安装的包的大小是正确的。这是207兆字节的Nvidia-驱动器-440。如果比较少的话，就意味着你还没有把所有的包都拿掉。|599248|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_20.04_LTS_(Focal_Fossa)|1|IMAGE|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/2d904c8217e217088e4e83c40c62a83a.jpg|imageAlt^0|0|0|0|4|C|0|0|0|0|0|1|1|0|0|0^^$0|@$1|2|3|4|5|6|7|1A|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|1B|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|1C|8|@]|9|@]|A|$]]|$1|I|3|J|5|6|7|1D|8|@]|9|@$K|1E|L|1F|1|1G]]|A|$]]|$1|M|3|N|5|6|7|1H|8|@]|9|@]|A|$]]|$1|O|3|P|5|6|7|1I|8|@]|9|@]|A|$]]|$1|Q|3|R|5|S|7|1J|8|@]|9|@$K|1K|L|1L|1|1M]]|A|$]]|$1|T|3|P|5|6|7|1N|8|@]|9|@]|A|$]]|$1|U|3|V|5|6|7|1O|8|@]|9|@]|A|$]]|$1|W|3|-4|5|6|7|1P|8|@]|9|@]|A|$]]]|X|$Y|$5|Z|10|11|A|$12|13]]|14|$5|15|10|16|A|$17|18|19|-4]]]]

I have to restart my kernels and remove all the packages that I have installed previously (during the first installation). Please make sure to delete all the packages, even after removing packages by the command below:
<pre class="lang-none prettyprint-override"><code>sudo apt-get --purge remove &quot;*nvidia*&quot;
</code></pre>
The packages, like &quot;libtinfo6:i386&quot;, don't get removed.
I'm using <a href="https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_20.04_LTS_(Focal_Fossa)" rel="nofollow noreferrer">Ubuntu 20.04</a> (Focal Fossa) and Nvidia-driver-440. For that, you have to remove all the packages shown in the below image.
List of all the packages that need to be remove:
<a href="https://i.stack.imgur.com/v1F4E.jpg" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/v1F4E.jpg" alt="img" /></a>
As shown in the image, make sure that the package you're installing is of the correct size. That is 207  MB for Nvidia-driver-440. If it's less, it means you haven't removed all the packages.

blocks|key|599266|text|为了完整起见，我也遇到了这个问题。在我的例子中，由于我将Clang设置为我的默认编译器(使用update-alternatives)，nvidia-driver-440无法编译(检查/var/crash/)，尽管apt没有发布任何警告。对我来说，解决方案是apt+purge+nvidia-*，将cc设置为使用gcc、重新启动和重新安装nvidia-driver-440。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|599267|entityMap^0|1A|J|1V|H|2J|B|2Y|3|3K|I|44|2|4B|3|4O|H|0^^$0|@$1|2|3|4|5|6|7|H|8|@$9|I|A|J|B|C]|$9|K|A|L|B|C]|$9|M|A|N|B|C]|$9|O|A|P|B|C]|$9|Q|A|R|B|C]|$9|S|A|T|B|C]|$9|U|A|V|B|C]|$9|W|A|X|B|C]]|D|@]|E|$]]|$1|F|3|-4|5|6|7|Y|8|@]|D|@]|E|$]]]|G|$]]

For completeness, I ran into this issue as well. In my case it turned out that because I had set Clang as my default compiler (using <code>update-alternatives</code>), <code>nvidia-driver-440</code> failed to compile (check <code>/var/crash/</code>) even though <code>apt</code> didn't post any warnings. For me, the solution was to <code>apt purge nvidia-*</code>, set <code>cc</code> back to use <code>gcc</code>, reboot, and reinstall <code>nvidia-driver-440</code>.

blocks|key|920126|text|大多数情况下，重启将修复Ubuntu+18.04+(仿生海狸)上的问题。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|920127|“未能初始化NVML:驱动程序/库版本错配？”错误通常意味着CUDA驱动程序仍在运行一个与当前使用的CUDA工具包版本不兼容的旧版本。重新启动计算节点通常会解决此问题。|920128|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_18.04_LTS_.28Bionic_Beaver.29^0|C|C|0|0|0^^$0|@$1|2|3|4|5|6|7|N|8|@]|9|@$A|O|B|P|1|Q]]|C|$]]|$1|D|3|E|5|6|7|R|8|@]|9|@]|C|$]]|$1|F|3|-4|5|6|7|S|8|@]|9|@]|C|$]]]|G|$H|$5|I|J|K|C|$L|M]]]]

Mostly reboot would fix the issue on <a href="https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_18.04_LTS_.28Bionic_Beaver.29" rel="nofollow noreferrer">Ubuntu 18.04</a> (Bionic Beaver).
The “Failed to initialize NVML: Driver/library version mismatch?” error generally means the CUDA Driver is still running an older release that is incompatible with the CUDA toolkit version currently in use. Rebooting the compute nodes will generally resolve this issue.

blocks|key|599305|text|首先我安装了Nvidia驱动程序。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|599306|接下来我安装了CUDA。|599307|在那之后，我得到了“驱动/库版本错配”错误，但是我可以看到CUDA版本，所以我清除了Nvidia驱动程序并重新安装它。|599308|然后它就正常工作了。|599309|entityMap^0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|J|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|K|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|L|8|@]|9|@]|A|$]]|$1|F|3|G|5|6|7|M|8|@]|9|@]|A|$]]|$1|H|3|-4|5|6|7|N|8|@]|9|@]|A|$]]]|I|$]]

First I installed the Nvidia driver.
Next I installed CUDA.
After that, I got the &quot;Driver/library version mismatch&quot; error, but I could see the CUDA version, so I purged the Nvidia driver and reinstalled it.
Then it worked correctly.

blocks|key|918166|text|有一个更简单的解决方案对我有效。在软呢帽+33上，尝试以下操作：|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|918167|rpm+-qa+%7C+grep+-i+nvidia+%7C+grep+f32|code-block|syntax|javascript|918168|您应该有两个包是从OpenGL的Fedora上一个版本中列出的。移除这些并重新启动。|918169|删除和重新安装整个Nvidia软件包集是过分的。|918170|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Fedora_%2528operating_system%2529|1|https://en.wikipedia.org/wiki/OpenGL^0|H|3|0|0|0|9|6|1|0|0^^$0|@$1|2|3|4|5|6|7|W|8|@]|9|@$A|X|B|Y|1|Z]]|C|$]]|$1|D|3|E|5|F|7|10|8|@]|9|@]|C|$G|H]]|$1|I|3|J|5|6|7|11|8|@]|9|@$A|12|B|13|1|14]]|C|$]]|$1|K|3|L|5|6|7|15|8|@]|9|@]|C|$]]|$1|M|3|-4|5|6|7|16|8|@]|9|@]|C|$]]]|N|$O|$5|P|Q|R|C|$S|T]]|U|$5|P|Q|R|C|$S|V]]]]

There is an easier solution that worked for me. On <a href="https://en.wikipedia.org/wiki/Fedora_%28operating_system%29" rel="nofollow noreferrer">Fedora</a> 33, try the following:
<pre class="lang-none prettyprint-override"><code>rpm -qa | grep -i nvidia | grep f32
</code></pre>
You should have two packages listed from the previous version of Fedora for <a href="https://en.wikipedia.org/wiki/OpenGL" rel="nofollow noreferrer">OpenGL</a>. Remove those and reboot.
Deleting and reinstalling the entire Nvidia package set is overkill.

blocks|key|2314005|text|我也面临着同样的问题，我在这里发布了我的解决方案。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2314006|在我的例子中，NVRM版本为440.100，驱动程序版本为460.32.03。我的驱动程序是由sudo+apt+install+caffe-cuda更新的，当时我没有注意到，但我从/var/log/apt/history.log上查看了它。|offset|length|style|CODE|2314007|通过遵循我的NVRM版本，我只使用了sudo+apt+install+nvidia-driver-440，但它安装了450.102。我不知道为什么它安装了另一个版本，nvidia-smi显示了450.102.04。|2314008|无论如何，在重新启动我的电脑后，现在一切都很好。重新安装驱动程序后，我的数据自动化系统仍然正常工作。|2314009|我没有删除/清除任何与Nvidia驱动程序相关的内容。版本460.32.03是通过运行sudo+apt+install+nvidia-driver-440自动卸载的。|BOLD|2314010|entityMap^0|0|1B|R|2I|O|0|I|Y|0|0|3|5|17|Y|0^^$0|@$1|2|3|4|5|6|7|Q|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|R|8|@$D|S|E|T|F|G]|$D|U|E|V|F|G]]|9|@]|A|$]]|$1|H|3|I|5|6|7|W|8|@$D|X|E|Y|F|G]]|9|@]|A|$]]|$1|J|3|K|5|6|7|Z|8|@]|9|@]|A|$]]|$1|L|3|M|5|6|7|10|8|@$D|11|E|12|F|N]|$D|13|E|14|F|G]]|9|@]|A|$]]|$1|O|3|-4|5|6|7|15|8|@]|9|@]|A|$]]]|P|$]]

I was facing the same problem and I'm posting my solution here.
In my case, the NVRM version was 440.100 and the driver version was 460.32.03. My driver was updated by <code>sudo apt install caffe-cuda</code> and I didn't notice at that time, but I checked it from <code>/var/log/apt/history.log</code>.
By following my NVRM version, I just used <code>sudo apt install nvidia-driver-440</code>, but it installed 450.102. I don't know why it installed another version and nvidia-smi is showing 450.102.04.
Anyhow, after rebooting my PC, everything is working fine now. After reinstalling the driver, still my CUDA is working fine.
I didn't remove/purge anything related to the Nvidia driver. Version 460.32.03 was uninstalled automatically by running <code>sudo apt install nvidia-driver-440</code>.

blocks|key|2314026|text|为什么会发生版本错配，我们如何防止它再次发生？|type|unstyled|depth|inlineStyleRanges|entityRanges|data|2314027|您可能会发现，在这些位置上，nvidia-*的版本不同：|offset|length|style|CODE|2314028|dpkg+-l+%7C+grep+nvidia+(查看nvidia-utils-xxx包版本)，以及|ordered-list-item|2314029|cat+/proc/driver/nvidia/version+(查看内核模块的版本，460.56+-例如)|2314030|重新启动应该有效，但是您可能希望通过修改/etc/apt/sources.list.d/文件来禁止该包的自动更新，或者通过执行命令apt-mark+hold+nvidia-utils-version_number来简单地对包进行保持。|2314031|P.S.：一些内容是受这的启发(原文是中文的，所以我引用了翻译的版本)。|2314032|entityMap|0|LINK|mutability|MUTABLE|url|https://askubuntu.com/questions/18654/how-to-prevent-updating-of-a-specific-package|1|https://www.programmersought.com/article/29283744928/^0|0|E|8|0|0|L|P|G|0|0|V|17|6|0|K|O|1T|15|36|2|0|0|B|1|1|0^^$0|@$1|2|3|4|5|6|7|10|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|11|8|@$D|12|E|13|F|G]]|9|@]|A|$]]|$1|H|3|I|5|J|7|14|8|@$D|15|E|16|F|G]|$D|17|E|18|F|G]]|9|@]|A|$]]|$1|K|3|L|5|J|7|19|8|@$D|1A|E|1B|F|G]|$D|1C|E|1D|F|G]]|9|@]|A|$]]|$1|M|3|N|5|6|7|1E|8|@$D|1F|E|1G|F|G]|$D|1H|E|1I|F|G]]|9|@$D|1J|E|1K|1|1L]]|A|$]]|$1|O|3|P|5|6|7|1M|8|@]|9|@$D|1N|E|1O|1|1P]]|A|$]]|$1|Q|3|-4|5|6|7|1Q|8|@]|9|@]|A|$]]]|R|$S|$5|T|U|V|A|$W|X]]|Y|$5|T|U|V|A|$W|Z]]]]

Why does the version mismatch happen and how can we prevent it from happening again?
You may find that the versions of <code>nvidia-*</code> are different in these locations:
<ol>
<li><code>dpkg -l | grep nvidia</code> (look at <code>nvidia-utils-xxx</code> package
version), and</li>
<li><code>cat /proc/driver/nvidia/version</code> (look at the version of Kernel
Module, <code>460.56</code> - for example)</li>
</ol>
The restart should work, but you may want to forbid the automatic update of this package by modifying <code>/etc/apt/sources.list.d/</code> files or simply <a href="https://askubuntu.com/questions/18654/how-to-prevent-updating-of-a-specific-package">hold</a> the package by executing the command <code>apt-mark hold nvidia-utils-version_number</code>.
P.S.: Some content was inspired by <a href="https://www.programmersought.com/article/29283744928/" rel="noreferrer">this</a> (the original instruction was in Chinese, so I referenced the translated version instead)

blocks|key|920205|text|重新启动或卸载驱动程序对我不起作用。我通过更新我的Nvidia驱动器440.33.01到450.80.2来解决这个问题。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|920206|sudo+apt-get+install+nvidia-driver-450

sudo+reboot|code-block|syntax|javascript|920207|我正在运行Ubuntu+20.04+LTS+(焦距Fossa)，这是一个远程服务器。|offset|length|920208|entityMap|0|LINK|mutability|MUTABLE|url|https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_20.04_LTS_(Focal_Fossa)^0|0|0|5|C|0|0^^$0|@$1|2|3|4|5|6|7|S|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|T|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|U|8|@]|9|@$I|V|J|W|1|X]]|A|$]]|$1|K|3|-4|5|6|7|Y|8|@]|9|@]|A|$]]]|L|$M|$5|N|O|P|A|$Q|R]]]]

It doesn't work for me by rebooting or unloading the driver.
I solved the problem by updating my Nvidia driver 440.33.01 to 450.80.2.
<pre class="lang-none prettyprint-override"><code>sudo apt-get install nvidia-driver-450

sudo reboot
</code></pre>
I'm running <a href="https://en.wikipedia.org/wiki/Ubuntu_version_history#Ubuntu_20.04_LTS_(Focal_Fossa)" rel="nofollow noreferrer">Ubuntu 20.04</a> LTS (Focal Fossa), which is a remote server.

When I run <code>nvidia-smi</code>, I get the following message:
<blockquote>
Failed to initialize NVML: Driver/library version mismatch
</blockquote>
An hour ago I received the same message and uninstalled my CUDA library and I was able to run <code>nvidia-smi</code>, getting the following result:
<a href="https://i.stack.imgur.com/RyAeQ.png" rel="noreferrer"><img src="https://i.stack.imgur.com/RyAeQ.png" alt="nvidia-smi-result" /></a>
After this I downloaded <code>cuda-repo-ubuntu1604-8-0-local-ga2_8.0.61-1_amd64.deb</code> from <a href="https://developer.nvidia.com/cuda-downloads" rel="noreferrer">the official NVIDIA page</a> and then simply:
<pre class="lang-none prettyprint-override"><code>sudo dpkg -i cuda-repo-ubuntu1604-8-0-local-ga2_8.0.61-1_amd64.deb
sudo apt-get update
sudo apt-get install cuda
export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}
</code></pre>
Now I have CUDA installed, but I get the mentioned mismatch error.
<hr />
Some potentially useful information:
Running <code>cat /proc/driver/nvidia/version</code> I get:
<pre class="lang-none prettyprint-override"><code>NVRM version: NVIDIA UNIX x86_64 Kernel Module 378.13 Tue Feb 7 20:10:06 PST 2017
GCC version: gcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.4)
</code></pre>
I'm running Ubuntu 16.04.2 LTS (Xenial Xerus).
The kernel release is 4.4.0-66-generic.

Nvidia NVML Driver/library version mismatch

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

教程

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云智能顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

EdgeOne AI 安全实战专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋

腾讯云代码助手

云原生构建

TAPD 敏捷项目管理

Cloud Studio

SDK中心

API中心

命令行工具

涵盖代码开发、场景应用、自动测试全流程，助你从零构建专属AI助手

一站式MCP教程库，解锁AI应用新玩法

聚焦“写作效率、视觉美观与运行性能”三方面进行全面升级，为您提供更高效、稳定的创作环境

社区富文本&Markdown编辑器全新改版上线，欢迎大家体验!

诚挚邀请您参与本次调研，分享您的真实使用感受与建议。您的反馈至关重要，感谢您的支持与参与！

社区新版编辑器体验调研

当我运行nvidia-smi时，我会收到以下消息：未能初始化NVML:驱动程序/库版本不匹配一个小时前，我收到了同样的消息，并卸载了我的CUDA库，并且能够运行nvidia-smi，得到以下结果：​​在此之后，我从下载了，然后简单地：sudo dpkg -i cuda-repo-ubuntu1604-8-0-local...

问NVML驱动程序/库版本错配
EN

Stack Overflow用户

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问NVML驱动程序/库版本错配EN