Skip to content

Project

LiteLLM

LiteLLM is an open-source gateway/proxy that routes calls to many LLM providers through one API, offering logging, spend tracking, and self-hosted deployment.

Known aliases

  • BerriAI LiteLLM
  • BerriAI/litellm
  • Berri LiteLLM
  • ghcr.io/berriai/litellm
  • litellm
  • LiteLLM-class gateways
  • LiteLLM gateway
  • LiteLLM proxy
  • LiteLLM Python v1
  • LiteLLM Rust beta

Current stories

buildOne report1 publisher

Check Point finds two LiteLLM flaws that expose every provider key the gateway stores

LiteLLM 1.83.9 fixes two Check Point-reported flaws, CVE-2026-35029 and CVE-2026-59823, that affect every older version of the gateway. A DEV Community patch guide pairs the upgrade with key rotation and an audit of what the gateway held.

Publishers:dev.to

Reality

Evidence45
Adoption55
Hype gap+20
Incentives
Not scored
Confidence50
securityConfirmed3 publishers

Two LiteLLM flaws let a basic user account take over the gateway or capture AI provider keys

Check Point Research disclosed two LiteLLM flaws, CVE-2026-35029 and CVE-2026-59823, both fixed in version 1.83.9. On older versions a basic account is enough to take over the gateway or capture provider API keys, according to Check Point's testing.

Perspective Coverage

3 publishers
Builder
Builder 32%
Operator
Operator 51%
Investor
Investor 17%

Reality

Evidence58
Adoption70
Hype gap+15
Incentives50
Confidence65
securityConfirmed5 publishers

PoeLLM botnet has turned more than 2,100 exposed servers into miners and scanners

Lumen's Black Lotus Labs says the PoeLLM cryptomining botnet has compromised more than 2,100 servers, many running exposed LiteLLM or Ollama. Each host it takes then mines and scans for the next, so an AI server left open to the internet is attack surface to lock down.

Perspective Coverage

6 publishers
Builder
Builder 34%
Operator
Operator 59%
Investor
Investor 7%

Reality

Evidence70
Adoption50
Hype gap+5
Incentives
Not scored
Confidence70
securityConfirmed16 publishers

AFP sizes the Shai-Hulud syndicate's take at more than 500,000 credentials

Two men arrested in Australia this week are alleged TeamPCP members. The AFP estimate filed alongside the case puts the worm's take at more than 500,000 credentials and 300GB of data, and that scale makes token lifetime the live question.

Perspective Coverage

16 publishers
Builder
Builder 29%
Operator
Operator 56%
Investor
Investor 15%

Reality

Evidence62
Adoption68
Hype gap+30
Incentives58
Confidence70
securityConfirmed4 publishers

CISA's seven new KEV entries put SonicWall gateways and Artifactory on one patch clock

SonicWall's CVSS 10.0 SSRF chains into command execution on remote access appliances, and JFrog Artifactory hands unauthenticated attackers admin under default configuration. Reverse shells and miners are already landing.

Perspective Coverage

4 publishers
Builder
Builder 14%
Operator
Operator 72%
Investor
Investor 14%

Reality

Evidence70
Adoption
Not scored
Hype gap+10
Incentives35
Confidence68
securityOne report1 publisher

Pwn2Own Ireland's first day puts LiteLLM and OpenAI Codex among 32 exploited zero-days

Pwn2Own Ireland 2026 researchers exploited 32 zero-days on day one for $388,500, including LiteLLM flaws and an OpenAI Codex argument injection. Vendors now have 90 days to ship fixes before ZDI publishes details, and the report describes contest exploits only.

Reality

Evidence55
Adoption
Not scored
Hype gap+10
Incentives35
Confidence55
securityOne report1 publisher

Pwn2Own Ireland's first-day draw sends three teams at the Galaxy S26 and two at LiteLLM

ZDI's Pwn2Own Ireland 2026 schedule sets 21 exploit attempts in Cork on October 6, the largest a $300,000 remote Pixel 10 entry. The schedule names exact models and time slots, so operators of the targeted products know which results to look for on ZDI's live leaderboard.

Publishers:thezdi.com

Reality

Evidence62
Adoption
Not scored
Hype gap+5
Incentives40
Confidence65
buildOne report1 publisher

MCP Atlassian hands unidentified HTTP callers the operator's Jira and Confluence access

MCP Atlassian releases before 0.22.0 fall back to the operator's credentials whenever an HTTP caller has no verified identity (CVE-2026-77244). HTTP deployments need the upgrade, and the endpoint should be reachable only through a management network or an authenticating proxy.

Publishers:dev.to

Reality

Evidence40
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence40
buildOne report1 publisher

MCP's Python SDK fix needs a hard-coded OAuth issuer to protect machine-to-machine clients

MCP Python SDK maintainers rated a flaw that let a connected server choose where OAuth secrets were sent High, at 7.5. For its two machine-to-machine providers, upgrading changes nothing until the client names the issuer it expects.

Publishers:dev.to

Reality

Evidence50
Adoption
Not scored
Hype gap+15
Incentives30
Confidence50
buildOne report1 publisher

Four MCP servers drew CVEs for exposing every tool without authentication

NVD logged CVEs for four MCP servers in about 35 hours, each because every tool it exposes needs no authentication. A fifth MCP flaw, LiteLLM's authentication bypass, is already on CISA's exploited-vulnerabilities list.

Publishers:dev.to

Reality

Evidence62
Adoption58
Hype gap−6
Incentives45
Confidence52
buildOne report1 publisher

LiteLLM's standalone Rust gateway adds 0.7 ms at p99 with logging and spend tracking switched off

LiteLLM's standalone Rust path added 0.7 ms of p99 latency against 257.7 ms for its Python proxy in July 22 benchmark artifacts reviewed on dev.to. Logging, persistence and spend tracking were off for the run, so the gap measures forwarding alone.

Publishers:dev.to

Reality

Evidence35
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence40
buildOne report1 publisher

LiteLLM's Lens moves agent-failure analysis into the self-hosted gateway that routes model calls

LiteLLM launched Lens on September 30, a tool that uses AI agents to find recurring failures across agent traces sent through its model gateway. Customers host the analyzer and its databases, and the 200,000-trace volume CTO Ishaan Jaffer cites is a future target Lens has not been measured against.

Publishers:runtimewire.com

Reality

Evidence40
Adoption
Not scored
Hype gap+25
Incentives60
Confidence40
securityConfirmed4 publishers

Google traces most of 2026's exploitation growth to fast n-day weaponization

Google's threat intelligence group counts 18 exploited flaws a month in 2026, up from 10.5 in 2025, while zero-days rose only from eight to 11. GTIG attributes most of the added attacks to fast weaponization of disclosed n-days, so the exposure sits in the days after a patch ships.

Perspective Coverage

4 publishers
Builder
Builder 33%
Operator
Operator 61%
Investor
Investor 6%

Reality

Evidence72
Adoption
Not scored
Hype gap+30
Incentives35
Confidence65
productOne report1 publisher

Attackers exploited one in 431 new vulnerabilities this year by Google's count

Google's Threat Intelligence Group counted 141 flaws exploited in the wild from January to August, while monthly disclosures doubled to 10,740. Patch teams do better sorting by that exploited set than by the total, though attackers now reach some public flaws within days.

Publishers:siliconangle.com

Reality

Evidence55
Adoption
Not scored
Hype gap+15
Incentives60
Confidence50
buildOne report1 publisher

A low-privilege LiteLLM key could run system commands through its MCP test endpoints

LiteLLM's MCP test endpoints let any valid proxy key run arbitrary commands on the gateway, rated CVSS 8.8. CISA added the flaw to its Known Exploited Vulnerabilities catalog on June 8, 2026, confirming exploitation in the wild.

Publishers:dev.to

Reality

Evidence50
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence50
buildOne report1 publisher

Hosted agent sandboxes from OpenAI, Perplexity and Gobare disagree on what survives idle time

Misha, who works on Gobare, found in public docs that OpenAI may delete a sandbox idle for an hour and Gobare pauses one after five minutes. Perplexity promises nothing between responses, so jobs that wait on a human or serve a preview have to fit one of those clocks.

Publishers:dev.to

Reality

Evidence45
Adoption
Not scored
Hype gap+10
Incentives60
Confidence50

Earlier coverage

  1. Whoever holds a LiteLLM master key reads every provider key and the host's IAM credentials

    Security · September 10, 2026 · Confirmed2 publishers