<?xml version='1.0' encoding='UTF-8'?>
<?xml-stylesheet href="/static/style.xsl" type="text/xsl"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" version="2.0">
  <channel>
    <title>Most recent entries from all</title>
    <link>https://cve.radiocsirt.org</link>
    <description>Contains only the most 10 recent entries.</description>
    <docs>http://www.rssboard.org/rss-specification</docs>
    <generator>python-feedgen</generator>
    <language>en</language>
    <lastBuildDate>Sat, 03 Oct 2026 04:02:34 +0000</lastBuildDate>
    <item>
      <title>CVE-2026-34753 — vLLM affected by Server-Side Request Forgery (SSRF) in `download_bytes_from_url `</title>
      <link>https://cve.radiocsirt.org/vuln/cve-2026-34753</link>
      <description>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; vllm-project vllm&lt;/p&gt;
&lt;p&gt;vLLM is an inference and serving engine for large language models (LLMs). From 0.16.0 to before 0.19.0, a server-side request forgery (SSRF) vulnerability in download_bytes_from_url allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions.
This can be used to target internal services (e.g. cloud metadata endpoints or internal HTTP APIs) reachable from the vLLM host. This vulnerability is fixed in 0.19.0.&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; vllm-project vllm&lt;/p&gt;
&lt;p&gt;vLLM is an inference and serving engine for large language models (LLMs). From 0.16.0 to before 0.19.0, a server-side request forgery (SSRF) vulnerability in download_bytes_from_url allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions.
This can be used to target internal services (e.g. cloud metadata endpoints or internal HTTP APIs) reachable from the vLLM host. This vulnerability is fixed in 0.19.0.&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://cve.radiocsirt.org/vuln/cve-2026-34753</guid>
    </item>
    <item>
      <title>GHSA-pf3h-qjgv-vcpr — vLLM: Server-Side Request Forgery (SSRF) in `download_bytes_from_url `</title>
      <link>https://cve.radiocsirt.org/vuln/ghsa-pf3h-qjgv-vcpr</link>
      <description>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; PyPI: vllm&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;A Server Side Request Forgery (SSRF) vulnerability in `download_bytes_from_url` allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions.&lt;/p&gt;
&lt;p&gt;This can be used to target internal services (e.g. cloud metadata endpoints or internal HTTP APIs) reachable from the vLLM host.&lt;/p&gt;
&lt;p&gt;------&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;#### Vulnerable component&lt;/p&gt;
&lt;p&gt;The vulnerable logic is in the batch runner entrypoint `vllm/entrypoints/openai/run_batch.py`, function `download_bytes_from_url`:&lt;/p&gt;
&lt;p&gt;```
# run_batch.py Lines 442-482
async def download_bytes_from_url(url: str) -&amp;gt; bytes:
    &amp;#34;&amp;#34;&amp;#34;
    Download data from a URL or decode from a data URL.&lt;/p&gt;
&lt;p&gt;Args:
        url: Either an HTTP/HTTPS URL or a data URL (data:...;base64,...)&lt;/p&gt;
&lt;p&gt;Returns:
        Data as bytes
    &amp;#34;&amp;#34;&amp;#34;
    parsed = urlparse(url)&lt;/p&gt;
&lt;p&gt;# Handle data URLs (base64 encoded)
    if parsed.scheme == &amp;#34;data&amp;#34;:
        # Format: data:...;base64,&amp;lt;base64_data&amp;gt;
        if &amp;#34;,&amp;#34; in url:
            header, data = url.split(&amp;#34;,&amp;#34;, 1)
            if &amp;#34;base64&amp;#34; in header:
                return base64.b64decode(data)
            else:
                raise ValueError(f&amp;#34;Unsupported data URL encoding: {header}&amp;#34;)
        else:
            raise ValueError(f&amp;#34;Invalid data URL format: {url}&amp;#34;)&lt;/p&gt;
&lt;p&gt;# Handle HTTP/HTTPS URLs
    elif parsed.scheme in (&amp;#34;http&amp;#34;, &amp;#34;https&amp;#34;):
        async with (
            aiohttp.ClientSession() as session,
            sessi…&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; PyPI: vllm&lt;/p&gt;
&lt;p&gt;### Summary&lt;/p&gt;
&lt;p&gt;A Server Side Request Forgery (SSRF) vulnerability in `download_bytes_from_url` allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions.&lt;/p&gt;
&lt;p&gt;This can be used to target internal services (e.g. cloud metadata endpoints or internal HTTP APIs) reachable from the vLLM host.&lt;/p&gt;
&lt;p&gt;------&lt;/p&gt;
&lt;p&gt;### Details&lt;/p&gt;
&lt;p&gt;#### Vulnerable component&lt;/p&gt;
&lt;p&gt;The vulnerable logic is in the batch runner entrypoint `vllm/entrypoints/openai/run_batch.py`, function `download_bytes_from_url`:&lt;/p&gt;
&lt;p&gt;```
# run_batch.py Lines 442-482
async def download_bytes_from_url(url: str) -&amp;gt; bytes:
    &amp;#34;&amp;#34;&amp;#34;
    Download data from a URL or decode from a data URL.&lt;/p&gt;
&lt;p&gt;Args:
        url: Either an HTTP/HTTPS URL or a data URL (data:...;base64,...)&lt;/p&gt;
&lt;p&gt;Returns:
        Data as bytes
    &amp;#34;&amp;#34;&amp;#34;
    parsed = urlparse(url)&lt;/p&gt;
&lt;p&gt;# Handle data URLs (base64 encoded)
    if parsed.scheme == &amp;#34;data&amp;#34;:
        # Format: data:...;base64,&amp;lt;base64_data&amp;gt;
        if &amp;#34;,&amp;#34; in url:
            header, data = url.split(&amp;#34;,&amp;#34;, 1)
            if &amp;#34;base64&amp;#34; in header:
                return base64.b64decode(data)
            else:
                raise ValueError(f&amp;#34;Unsupported data URL encoding: {header}&amp;#34;)
        else:
            raise ValueError(f&amp;#34;Invalid data URL format: {url}&amp;#34;)&lt;/p&gt;
&lt;p&gt;# Handle HTTP/HTTPS URLs
    elif parsed.scheme in (&amp;#34;http&amp;#34;, &amp;#34;https&amp;#34;):
        async with (
            aiohttp.ClientSession() as session,
            sessi…&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://cve.radiocsirt.org/vuln/ghsa-pf3h-qjgv-vcpr</guid>
    </item>
  </channel>
</rss>
