LiteLLM
Open-source Python SDK and proxy gateway exposing many LLM providers through an OpenAI-style API.
HN commenters use it as a proxy that puts multiple LLM providers behind one API base1234 for coding assistants and other tools, and for monitoring usage.5678 The most common complaint is supply-chain security after a credential-stealer incident in its builds,9101112 and others report flakiness and outages.1314
Why people recommend it
Unified LLM proxy · 3
Lets tools like Claude Code and vim-ai reach different model providers through one proxy endpoint.
“I use a LiteLLM proxy with Claude Code”
“I use litellm (slightly modified to allow cloud code telemetry pass through) and langfuse.”
Show 1 more
“through vim-ai and liteLLM”
Usage monitoring · 3
Used as a bridge to monitor AI usage and record traffic for observability or auditing.
“litellm as ai bridge to monitor usage”
“I use litellm (slightly modified to allow cloud code telemetry pass through) and langfuse.”
Show 1 more
“Aegis and LiteLLM, for example, are pre-execution firewalls that add a cryptographic audit trail.”
Internal rollout · 1
One company admin cancelled individual plans and rolled it out internally for staff.
“We just cancelled everyone's plans and rolled liteLLM out internally.”
Building block · 2
Suggested as a base layer to build on, and one commenter was glad to have used it.
“I would suggest to code it on top of ollama or LiteLLM”
“I'm glad I used litellm in my last project”
Caveats and dissent
Supply-chain security · 3
A credential-stealer incident in builds led some commenters to block it as a dependency or leave it.
“Announcing that there's a private enterprise repo you should be paying for if you don't want a credential stealer in your build is quite the optic”
“I'd like just flatout block litellm so I don't have to worry about picking it up as transitive dependency.”
Show 1 more
“This is the 2nd time. I'm glad I moved away from this platform.”
Enterprise-first fixes · 1
A fix for a further vulnerability was distributed first through a private registry for enterprise customers.
“Announcing that there's a private enterprise repo you should be paying for if you don't want a credential stealer in your build is quite the optic”
Reliability · 2
Some commenters describe it as flaky, outage-prone or poorly built.
“LiteLLM was incredible flaky and kept causing outages”
“LiteLLM and LangChain are AWFUL pieces of software and should be avoided at ALL costs.”
Setup friction · 1
Getting Claude Code running with Copilot APIs through the proxy was called a pain.
“I got Claude Code running with Copilot APIs via the LiteLLM proxy, but it was a pain in the butt.”
Routing latency · 1
One user noticed a significant delay in the routing process.
“I noticed a significant delay in the routing process”
Needs sandboxing · 1
A commenter advises sandboxing it separately from agents with minimal access.
“You sandbox LiteLLM separately from your agents, and you give it access to much less.”
Often compared with
Who recommends it in their own words
Threads where it came up
- Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor 4 mentions
- Ask HN: Who wants to be hired? (October 2026) 1 mention
- Ask HN: What do you run on a $5 VPS that's worth keeping online 24/7? 1 mention
- Should I reconsider my decision of self-hosting LLM Gateway? 1 mention
- Litelm: LiteLLM Without the Bloat 1 mention
- Show HN: We built open OpenRouter that turns usage into a better model 1 mention
Data: litellm.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 20
“I'd like just flatout block litellm so I don't have to worry about picking it up as transitive dependency.”
“You sandbox LiteLLM separately from your agents, and you give it access to much less.”
“This is the 2nd time. I'm glad I moved away from this platform.”
“Announcing that there's a private enterprise repo you should be paying for if you don't want a credential stealer in your build is quite the optic”
“LiteLLM model routing”
“litellm as ai bridge to monitor usage”
“was using LiteLLM for the company i work for. we were doing a poc.”
“LiteLLM and LangChain are AWFUL pieces of software and should be avoided at ALL costs.”
“to compare it with these other products that seem to be competing directly”
“I noticed a significant delay in the routing process”
“offline estimate priced against LiteLLM”
“Aegis and LiteLLM, for example, are pre-execution firewalls that add a cryptographic audit trail.”
“We just cancelled everyone's plans and rolled liteLLM out internally.”
“I'm glad I used litellm in my last project”
“I use a LiteLLM proxy with Claude Code”
“LiteLLM was incredible flaky and kept causing outages”
“I use litellm (slightly modified to allow cloud code telemetry pass through) and langfuse.”
“I got Claude Code running with Copilot APIs via the LiteLLM proxy, but it was a pain in the butt.”
“through vim-ai and liteLLM”
“I would suggest to code it on top of ollama or LiteLLM”