LiteLLM vs vLLM
LiteLLM vs vLLM: side-by-side comparison of two inference platforms & gateways systems — architecture, taxonomy, license, pricing, MCP/A2A support, and direct edges.
How they relate
- LiteLLM integrates with vLLM — LiteLLM docs have a vLLM provider page.
Cost & capability
| LiteLLM | vLLM | |
|---|---|---|
| Cost tier | searched not found | searched not found |
| $/Mtok input | searched not found | searched not found |
| $/Mtok output | searched not found | searched not found |
Where they differ (9)
Rows where both sides have data and the values disagree — the shortlist of dimensions that actually distinguish these two systems.
| LiteLLM | vLLM | |
|---|---|---|
| Type | Universal LLM API + Proxy | Fast LLM inference engine |
| Latest release | v1.106.4 (2026-10-25) | v0.32.0 (2026-10-18) |
| License | MIT (except enterprise/ directory, commercial license) | Apache-2.0 |
| GitHub | BerriAI/litellm — ~62.1k stars, last commit 2026-10 | vllm-project/vllm — ~95.4k stars, last commit 2026-10 |
| Pricing | OSS free; Enterprise custom pricing (30-day self-hosted trial key) | Free (open source, Apache-2.0) |
| Funding | BerriAI raised $1.6M seed 2023 (Y Combinator W23); commercial steward of LiteLLM OSS. | searched not found |
| Deployment | Self-hosted only — SDK and proxy; Enterprise self-hosted or air-gapped | Self-hosted only — pip / Docker inference server with OpenAI-compatible API; managed hosting only via third-party clouds |
| API surface | Python SDK, OpenAI-compatible proxy REST API | searched not found |
| MCP | yes — MCP Gateway in LiteLLM Proxy (Streamable HTTP, SSE, stdio) with per-key/team access control | searched not found |
At a glance
| LiteLLM | vLLM | |
|---|---|---|
| Section | Inference platforms & gateways | Inference platforms & gateways |
| Tier | T1 | T1 |
| Type | Universal LLM API + Proxy | Fast LLM inference engine |
| Created | 2023 | 2023 |
| Latest release | v1.106.4 (2026-10-25) | v0.32.0 (2026-10-18) |
| License | MIT (except enterprise/ directory, commercial license) | Apache-2.0 |
| GitHub | BerriAI/litellm — ~62.1k stars, last commit 2026-10 | vllm-project/vllm — ~95.4k stars, last commit 2026-10 |
| Pricing | OSS free; Enterprise custom pricing (30-day self-hosted trial key) | Free (open source, Apache-2.0) |
| Funding | BerriAI raised $1.6M seed 2023 (Y Combinator W23); commercial steward of LiteLLM OSS. | searched not found |
| Deployment | Self-hosted only — SDK and proxy; Enterprise self-hosted or air-gapped | Self-hosted only — pip / Docker inference server with OpenAI-compatible API; managed hosting only via third-party clouds |
| API surface | Python SDK, OpenAI-compatible proxy REST API | searched not found |
| Multi-tenancy | searched not found | searched not found |
| MCP | yes — MCP Gateway in LiteLLM Proxy (Streamable HTTP, SSE, stdio) with per-key/team access control | searched not found |
| A2A | searched not found | searched not found |
| OpenTelemetry | searched not found | searched not found |
| Optimised for | searched not found | searched not found |
| Anti-fit | searched not found | searched not found |
Pros & cons
LiteLLM
Pros: searched not found
Cons: searched not found
vLLM
Pros: searched not found
Cons: searched not found