LiteLLM vs Portkey
LiteLLM scores higher on the AgentReady, 84/100 against 63/100. They differ on 16 of the 41 signals. Which ones decides whether an agent can adopt them without a person watching.
What each one is
LiteLLM. The AI Gateway for platform teams.
Portkey. Portkey is a comprehensive platform designed to streamline and enhance AI integration for developers and organizations.
Where LiteLLM is ahead
LiteLLM passes llms-full.txt / full agent docs, and Portkey does not. That is discover, whether an agent can find the product at all without being told it exists.
It also holds understand: structured api reference, openapi / spec quality, response examples provided and errors and status codes documented. Portkey misses those.
And on adopt, programmatic credential creation, fast time to first request and copyable quickstart. Portkey misses those.
Finally, on operate, structured, predictable output, machine-readable errors, retry behavior documented and agent compatibility verified. Portkey misses those.
Where Portkey is ahead
Portkey passes clear canonical domain and clear product positioning, and LiteLLM does not. That is discover, whether an agent can find the product at all without being told it exists.
It also holds adopt: no mandatory sales call and agent-compatible signup flow. LiteLLM misses those.
What neither does
Both fail request examples provided, idempotency support, rate-limit behavior predictable. If your agent needs any of those, you will be building it yourself either way.
Score, pillar by pillar
The AgentReady splits into four pillars, scored separately, because a product can be easy to find and still impossible to adopt.
Discover. Portkey leads 93 to 80. LiteLLM misses clear canonical domain, clear product positioning; Portkey misses llms-full.txt / full agent docs.
Understand is whether an agent can read the docs and work out how the API behaves before calling it. LiteLLM leads 92 to 38. LiteLLM misses request examples provided; Portkey misses structured api reference, openapi / spec quality, request examples provided, response examples provided, errors and status codes documented.
Adopt. Both sit at 80/100 here. LiteLLM misses no mandatory sales call, agent-compatible signup flow; Portkey misses programmatic credential creation, fast time to first request, copyable quickstart.
Operate. LiteLLM leads 82 to 39. LiteLLM misses idempotency support, rate-limit behavior predictable; Portkey misses structured, predictable output, machine-readable errors, retry behavior documented, idempotency support, rate-limit behavior predictable, agent compatibility verified.
Pricing
LiteLLM starts at $0 and has a free tier. Portkey starts at $0/mo and has a free tier.
| LiteLLM plans | Portkey plans |
|---|---|
| Open Source $0 | Free $0/mo |
| Enterprise Talk to sales | - |
Signal by signal
| Signal | LiteLLM | Portkey |
|---|---|---|
| AgentReady | 84 | 63 |
| Discovery | 80 | 93 |
| Understanding | 92 | 38 |
| Adoption | 80 | 80 |
| Operability | 82 | 39 |
| Public API | Yes | Yes |
| MCP server | Yes | Yes |
| OpenAPI spec | Yes | Unknown |
| CLI | Yes | Yes |
| llms.txt | Yes | Yes |
| Self-serve signup | Yes | Yes |
| Free tier | Yes | Yes |
Which to pick
LiteLLM clears more of the signals an agent needs, so it is the safer default for unattended use. Full profiles: LiteLLM and Portkey. Alternatives to each: LiteLLM, Portkey.
An agent can fetch this as data: POST /v1/compare {"slugs": ["litellm", "portkey"]}