Automation foundation model for tiny devices: 2-bit, 8-29 MB, tool calls, ASR, structured extraction and embeddings on phones, wearables, smart homes, robots, cars and microcontrollers.
docs: correct --bits default description in README (default follows checkpoint bit map, not flat 4) (#69)
LeonSGPdc4ffd2b3be9
diff --git a/README.md b/README.mdindex 9843272..377100a 100644--- a/README.md+++ b/README.md@@ -121,7 +121,7 @@ pip install "cactus-needle[metal]"needle build checkpoints/needle2.pkl --lora checkpoints/needle_lora.pkl --out my_needle.cact```-Add `--bits 2` (default 4) for a smaller model, or set `NEEDLE_HF_REPO=<you>/<model>` and pass `--upload` to publish the `.cact`. The counterpart `needle download <you>/<model>/my_needle.cact` pulls a published archive on any machine.+Add `--bits 2` for a smaller model (by default the export follows the checkpoint's declared per-layer bit map, falling back to 4 when the checkpoint declares none), or set `NEEDLE_HF_REPO=<you>/<model>` and pass `--upload` to publish the `.cact`. The counterpart `needle download <you>/<model>/my_needle.cact` pulls a published archive on any machine.**4. Run it.** The engine is weights-agnostic, so a tuned `.cact` runs on it directly - no recompilation: