Model card: install and run the signed gbx_lm binary

The Python package is no longer published; the card now documents the notarised Apple Silicon binary, its macOS and chip requirements, and the memory this build actually takes.

README.md

@@ -4,7 +4,6 @@ language:
 - zh
 license: mit
 base_model: pipenetwork/GLM-5.3-Flash-MLX-4bit
-library_name: gbx-lm
 pipeline_tag: image-text-to-text
 tags:
 - mlx
@@ -27,12 +26,56 @@ of it.
 | `mtp/` | 3.90 GiB | draft head, off by default |
 
 Total 169.27 GiB. Of that, 165.37 GiB is the source build, whose bytes moved into
+
+These containers are not a format `mlx-lm` reads. The model runs on
+**`gbx_lm`**, a single signed binary for Apple Silicon; there is nothing to
+`pip install`.
 containers rather than being copied, and 3.90 GiB is the draft head, which no
 published build of this model carries.
 
-```python
-from gbx_lm.utils import load
-model, tokenizer = load("GreenBitAI/GLM-5.3-Flash-4bit-paged")
+## Requirements
+
+| | |
+| --- | --- |
+| macOS | **15.0 or later** |
+| chip | **Apple Silicon** (arm64). There is no Intel build. |
+| Python | none -- the binary carries what it needs |
+
+Memory is not a fixed figure for a paged build, and that is the point of one: it
+fills what fits and streams the rest from disk. On a 512 GB Mac Studio with room
+to spare this model settles at about **172 GB** resident. A smaller machine holds
+less and reads more from disk -- slower, but it runs.
+
+## Install
+
+```bash
+curl -fL -o gbx_lm-darwin-arm64.tar.gz 'https://github.com/GreenBitAI/gbx-lm/releases/latest/download/gbx_lm-darwin-arm64.tar.gz' \
+  && tar -xzf gbx_lm-darwin-arm64.tar.gz gbx_lm \
+  && mkdir -p "$HOME/.local/bin" \
+  && mv gbx_lm "$HOME/.local/bin/gbx_lm" \
+  && chmod +x "$HOME/.local/bin/gbx_lm"
+
+gbx_lm -h
+```
+
+`command not found` means `$HOME/.local/bin` is not on your `PATH`: add it, or
+call the binary by its full path. The build is signed with a Developer ID and
+notarised, so macOS runs it without the usual detour for a downloaded binary.
+
+## Run
+
+```bash
+gbx_lm --model GreenBitAI/GLM-5.3-Flash-4bit-paged --port 8111
+```
+
+That serves an OpenAI-compatible API on port 8111. The weights download on first
+use into `~/.libra/cache/models`; set `HF_HOME` to put them elsewhere, and
+`HF_TOKEN` if you meet the Hub's rate limits for anonymous downloads.
+
+```bash
+curl http://127.0.0.1:8111/v1/chat/completions \
+  -H 'Content-Type: application/json' \
+  -d '{"model":"GreenBitAI/GLM-5.3-Flash-4bit-paged","messages":[{"role":"user","content":"Hello"}]}'
 ```
 
 Where the weights fit they are filled from `experts.bin` and the model runs the
@@ -53,8 +96,8 @@ from `zai-org/GLM-5.3-Flash`, so speculative decoding works from this repository
 alone. It is off unless asked for:
 
 ```bash
-GBX_GLM53_MTP=on python -m gbx_lm.generate --model GreenBitAI/GLM-5.3-Flash-4bit-paged --max-tokens 256 --prompt "..."
-GBX_GLM53_MTP=on python -m gbx_lm.fastapi_server --model GreenBitAI/GLM-5.3-Flash-4bit-paged
+GBX_GLM53_MTP=on gbx_lm --model GreenBitAI/GLM-5.3-Flash-4bit-paged --port 8111
+GBX_GLM53_MTP=on gbx_lm --model GreenBitAI/GLM-5.3-Flash-4bit-paged --port 8111
 ```
 
 **It grows with the context.** Measured 2026-09-18 on a 512 GB Mac Studio