17 lines
558 B
Markdown
17 lines
558 B
Markdown
# Tools
|
|
|
|
These scripts support model preparation and offline engineering work. They are
|
|
not runtime dependencies of the C engine.
|
|
|
|
- `convert_fp8_to_int4.py`, `download_glm52.py`: model preparation
|
|
- `make_glm_oracle.py`, `make_glm_bench_model.py`: deterministic fixtures
|
|
- `benchmark_cuda_fixture.py`, `eval_glm.py`, `fetch_benchmarks.py`: benchmarks
|
|
- `gen_unicode.py`: tokenizer table generation
|
|
|
|
Run them from `c/`, for example:
|
|
|
|
```sh
|
|
python3 tools/convert_fp8_to_int4.py --selftest
|
|
python3 tools/make_glm_bench_model.py --output /tmp/colibri-bench
|
|
```
|