vLLM
docs · 5druntime · updated 2026-07-10
Specifications
| Language | python |
| Memory floor | — |
| Context Floor Tokens | 0 |
| Runner Kind | model |
| Os Support | Windows: wsl · Macos: limited · Linux: native |
| Supported API Formats | openai |
| Served API Formats | openai, anthropic |
| Cli Role | server |
| Flags | Default Server Runtime: false · Default Local Runtime For: |
Price components
| Type | USD | Basis | Source | Verified | Freshness |
|---|---|---|---|---|---|
| No published price. | |||||
Cost at three usage tiers
| A few questions and small tasks each day. | — |
| Daily use, plus some scheduled tasks. | — |
| Working for you most of the day, including web tasks. | — |
Fixed monthly cost; independent of usage.
What others don't publish
- vLLM's command line launches and administers its inference server; it is not an interactive chat CLI. Reach a vLLM deployment through its API endpoint. · docs.vllm.ai, 2026-07-10