Skip to content

Updated LiteLLM's Rust mode and agw performance blog - #895

Open
linsun wants to merge 2 commits into
mainfrom
lin-perf
Open

Updated LiteLLM's Rust mode and agw performance blog#895
linsun wants to merge 2 commits into
mainfrom
lin-perf

Conversation

@linsun

@linsun linsun commented Aug 13, 2026

Copy link
Copy Markdown
Contributor

an updated blog using LiteLLM's Rust mode.

linsun added 2 commits August 13, 2026 17:03
Signed-off-by: Lin Sun <lin.sun@solo.io>
Signed-off-by: Lin Sun <lin.sun@solo.io>
@cloudflare-workers-and-pages

cloudflare-workers-and-pages Bot commented Aug 13, 2026

Copy link
Copy Markdown

Deploying agentproxy with  Cloudflare Pages  Cloudflare Pages

Latest commit: eadee09
Status: ✅  Deploy successful!
Preview URL: https://9f128279.agentproxy.pages.dev
Branch Preview URL: https://lin-perf.agentproxy.pages.dev

View logs

@linsun linsun changed the title Updated litellm and agw performance blog Updated LiteLLM's Rust mode and agw performance blog Aug 13, 2026

### Visualized results

{{< reuse-image src="img/blog/agentgateway-vs-litellm-rust-mode/image1.png" width="624px" >}}

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Image 1 and 8, the y-axis doesn't seem to have numbers, says either 0 reqs/s or K reqs/s, not sure if it is missing a number before k, like maybe for image8 20K and 40K?

rust: true
```

I also made sure I was using the latest LiteLLM build available for the test, LiteLLM `1.98.0`, which is newer than `1.94.0`.

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

which is newer than 1.94.0.

not sure why the 1.94 is important to know? maybe say why or just remove

and then also, maybe add what version of agentgateway you were using?

Suggested change
I also made sure I was using the latest LiteLLM build available for the test, LiteLLM `1.98.0`, which is newer than `1.94.0`.
I also made sure I was using the latest LiteLLM build available for the test, LiteLLM `1.98.0`, and the latest agentgateway version, `1.4.1`.

| agentgateway | **35,502 QPS** | **0.863 ms** | **1.644 ms** | **1.972 ms** |
| LiteLLM (rust: true) | 984 QPS | 32.139 ms | 48.528 ms | 71.451 ms |

agentgateway handled **over 36× more requests per second** while maintaining sub-2 ms P99 latency.

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Typically we use sentence-case for agentgateway.

Suggested change
agentgateway handled **over 36× more requests per second** while maintaining sub-2 ms P99 latency.
Agentgateway handled **over 36× more requests per second** while maintaining sub-2 ms P99 latency.

Comment on lines +112 to +113
| agentgateway | 199% | 482% | **26 MB** | 34 MB |
| LiteLLM (rust: true) | 69% | 204% | **2.15 GB** | 2.15 GB |

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
| agentgateway | 199% | 482% | **26 MB** | 34 MB |
| LiteLLM (rust: true) | 69% | 204% | **2.15 GB** | 2.15 GB |
| agentgateway | 199% | 482% | **26 MiB** | 34 MiB |
| LiteLLM (rust: true) | 69% | 204% | **2.15 GiB** | 2.15 GiB |

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants