py313-roocode-code-indexer-macos (python/py-roocode-code-indexer-macos) Updated: 11 hours ago Add to my watchlist
High-performance OpenAI-compatible embedding server on Apple SiliconHigh-performance OpenAI-compatible embedding server on Apple Silicon. Runs Qwen3 embedding models locally via MLX with optimized batched GPU inference — no API keys needed. Up to 5x faster than Ollama for the same models.
Version: 0.1.2 License: MIT
GitHub
| Maintainers | No Maintainer |
| Categories | science python llm |
| Homepage | https://github.com/dmarkey/roocode-code-indexer-macos |
| Platforms | {darwin any} |
| Variants | - |
Subport(s) (3)
"py313-roocode-code-indexer-macos" depends on
lib (1)
run (6)
build (4)
Ports that depend on "py313-roocode-code-indexer-macos"
No ports
Port notes
Example MLX embedding server instance:
hf-313 download mlx-community/Qwen3-Embedding-8B-4bit-DWQ
caffeinate -i env HOST=localhost PORT=8080 \
roocode-code-indexer-macos-313 \
--model mlx-community/Qwen3-Embedding-8B-4bit-DWQ
curl -s http://localhost:8080/v1/embeddings \
-H 'Content-Type: application/json' \
-d '{"model": "mlx-community/Qwen3-Embedding-8B-4bit-DWQ", \
"input": ["cats purr", "kittens meow", "stock markets fell"]}' \
| python313 -c '
import sys, json, math
v = [d["embedding"] for d in json.load(sys.stdin)["data"]]
def cos(a, b):
return (sum(x*y for x, y in zip(a, b))
/ (math.sqrt(sum(x*x for x in a))
* math.sqrt(sum(y*y for y in b))))
print("cats vs kittens :", round(cos(v[0], v[1]), 3))
print("cats vs stocks :", round(cos(v[0], v[2]), 3))
print("kittens vs stocks:", round(cos(v[1], v[2]), 3))
'
Port Health:
Loading Port Health