mistral.rsvsvLLM
Two self-hosted LLM Inference projects, compared on the numbers selfhost.directory tracks live: score, popularity, release cadence and setup effort.
| mistral.rs | vLLM | |
|---|---|---|
| Score | 72/100 | 90/100 |
| Rank in category | #7 of 25 in LLM Inference | #3 of 25 in LLM Inference |
| GitHub stars | 7 842 | 97 011 |
| Stars, last 30 days | +101 | +3 341 |
| Stable releases, 12 months | 17 | 14 |
| Latest release | v0.9.5 · 19 Oct 2026 | v0.32.0 · 17 Oct 2026 |
| Setup | Moderate | Moderate |
| Official Docker image | Yes | Yes |
| License | MIT | Apache-2.0 |
| Main language | Rust | Python |
Which one should you pick?
vLLM scores 90/100 against 72/100 for mistral.rs, so it is the safer default.
mistral.rs ships more often: 17 stable releases in the past year, against 14 for vLLM.
vLLM is growing faster right now (+3 341 stars in 30 days).
Key differences
- License: mistral.rs is MIT, vLLM is Apache-2.0.
- Language: mistral.rs is written mainly in Rust, vLLM in Python.
- Platforms only mistral.rs lists: macOS, Windows.
- Features only mistral.rs lists: Prometheus Metrics.
- Features only vLLM lists: Continuous Batching, Disaggregated prefill/decode/encode, Distributed Inference, Embedding and retrieval models and 10 more.
mistral.rs details →vLLM details →All LLM Inference projects →