Rapid-MLX vs vllm-mlx: rag for Claude compared
data as of 2026-07-09
4.6k
combined stars
571
combined forks
13
ecosystem connections
100%
open-source
// summary
Rapid-MLX has more GitHub stars (3.2k vs 1.4k).
The right choice depends on your use case — the full spec is below.
Rapid-MLX
★ 3.2kactive
The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separa…
claudeapple-siliconclaude-codecursordeepseek
view full listing →
vllm-mlx
★ 1.4kactive
OpenAI and Anthropic compatible server for Apple Silicon. Run LLMs and vision-language models (Llama, Qwen-VL, LLaVA) with continuous batching, MCP tool call…
claudeanthropicapple-siliconaudio-processingclaude-code
view full listing →// head to head
| spec | Rapid-MLX | vllm-mlx |
|---|---|---|
| GitHub stars | 3.2k | 1.4k |
| Forks | 376 | 195 |
| Rating | — | — |
| Upvotes | 0 | 0 |
| Health | active | active |
| Last updated | 11 days ago | 16 days ago |
| Language | Python | Python |
| License | Apache-2.0 | Apache-2.0 |
| Pricing | open-source | open-source |
| Platforms | cli, api, desktop, web, mobile | cli, api, mobile |
| AI compatibility | claude | claude |
| Verified | no | no |
| Ecosystem connections | 10 | 3 |
More: all rag for Claude compared · the rag for Claude category
Own Rapid-MLX or vllm-mlx?
Claim your listing to keep its page accurate and earn CC.