From the experiment

Do coding agents recommend LiteLLM?

LiteLLM was chosen in 18% of 140 judged AI gateways sessions, ranking third. Measured with Claude Code, Codex and Cursor.

Published September 3, 2026 Read as Markdown

LiteLLM was chosen in 18% of 140 judged AI gateways sessions, ranking third. It was also raised as a candidate in 86 further sessions without being chosen.

This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in AI gateways inside a realistic codebase. Not what a chat assistant says about LiteLLM. What an agent actually installed.

The numbers

CategoryAI gateways
Sessions in the category140
Sessions where LiteLLM was chosen25
Install share18%
Rank in category3 of 8
Codebases it won in5
Raised as a candidate, not chosen86
Chosen when considered23%
Sitelitellm.ai

By agent

The three agents treat LiteLLM very differently. Codex chose it in 23% of its sessions in this category and Claude Code in 7%, a spread of 16 points.

AgentSessionsChose LiteLLMShare
Claude Code4537%
Codex481123%
Cursor471123%

A single blended install share would report the midpoint and hide both ends. If most of your users run Claude Code, your real position here is 7%, not 18%.

By who was asking

LiteLLM does much better with one kind of buyer than another. It won 29% of sessions asked as enterprise team and 0% of those asked as junior developer.

Who is askingSessionsChose LiteLLMShare
Vibe coder1600%
Junior developer3300%
Senior engineer561527%
Enterprise team351029%

What LiteLLM was up against

The full ranking in AI gateways, from the same sessions:

#ProductRuns wonShare
1Portkey3021%
2Cloudflare AI Gateway3021%
3LiteLLM (this page)2518%
4Vercel AI Gateway2216%
5OpenRouter1511%
6Built in-house (no product adopted)129%
7Amazon Bedrock32%
8Helicone11%

What this means

A solid second or third position in a category means the agent is genuinely choosing rather than reaching automatically. That is a winnable position, because the inputs it uses can be changed.

The fastest gains are usually in the sessions that were nearly won: read them, find the step where the agent moved on, and fix that step.

Where these numbers come from

The 140 sessions in AI gateways are part of a published set of 5,292, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.

Every AI gateways run can be replayed on the board.

If you work on LiteLLM: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.

<!-- generated by scripts/write-data-pages.mjs -->

Common questions

Do coding agents recommend LiteLLM?

Yes. LiteLLM was chosen in 25 of the 140 judged sessions in AI gateways, a 18% install share, ranking third in its category.

Does Claude Code recommend LiteLLM?

In 3 of the 45 sessions in AI gateways run with Claude Code, which is 7%.

Do different coding agents treat LiteLLM differently?

Yes, and by a wide margin. Codex chose it in 23% of its runs and Claude Code in 7%.

How was this measured?

Real coding agents at pinned versions were run in sandboxes inside 51 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.

How often is LiteLLM considered but not chosen?

It was raised as a candidate in 86 sessions without being chosen, and chosen in 25. That is a 23% conversion from considered to chosen.

Where this comes from

Armature ran 5,292 judged sessions with Claude Code, Codex and Cursor inside 51 realistic codebases, and published every run. The numbers on this page come from that work.

Read next

All library pages