saragossa

Fast pure-Rust Metal LLM inference engine for Apple Silicon: beats mlx_lm on MoE

A new way for AI to search for data

View on GitHub

8GitHub stars

Rust

What XERJ indexed

Snapshot

Whole repository

The whole repository at commit ae6c1b9e1a (2026-10-05).

Selected work

Destynova2
@Destynova2

“prendre en charge Qwen3.8 et détecter la tête MTP”. 15 files, 425 lines in src, src/serve, src/qwen_loader.

Licence

The repository

Apache-2.0

Checked by a person at the pinned commit.

The XERJ corpus

Apache-2.0, with credit

The author agreed by email on 2026-10-19. Same terms as the repository. Answers name the repo, the commit and the file:line.

in licence reviewnot in the hub yetsuggested use: adapt with attribution

Example questions

Where does saragossa handle “prendre en charge Qwen3.8 et détecter la tête MTP”?
What does src/serve do in saragossa?

The prompt

Paste it into your AI agent. It clones azerozero/saragossa at the pinned commit and indexes it on your machine.

Search the GitHub repository azerozero/saragossa with XERJ (https://hub.xerj.org/corpus/azerozero--saragossa).
Run these in a shell, in order, and check each result before the next:
1. xerj --help  (success: it prints usage). If xerj is missing: curl -fsSL https://xerj.org/get | sh
2. git clone https://github.com/azerozero/saragossa saragossa && git -C saragossa checkout ae6c1b9e1a097b60611f68cf4c22c9a6e1357e69
3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data
4. xerj autoindex ./saragossa --url http://127.0.0.1:9200 --prefix saragossa --progress plain
5. xerj autoindex map --url http://127.0.0.1:9200
Then answer my question from the indexed files and cite the ax_path of every hit. Query syntax: https://xerj.org/llms.txt
Licence: Apache-2.0. Keep the copyright notice and credit the authors.

Destynova2 AI, prompts and agents corpora

Selected work and consent: the XERJ registry.