Upload folder using huggingface_hub

README.md

@@ -25,15 +25,15 @@ line, so identifiers, paths and error text come through exactly as they were.
 
 - **1.2 MB.** 712 gradient-boosted trees, not a neural network.
 - **Milliseconds on one CPU core.** No GPU, no network call, no extra model request.
-- **Deterministic.** The same context and query give the same bytes on every machine.
+- **Deterministic.** No sampling and no randomness: the same context and query give the same output.
 - **Verbatim.** Every kept line is a line of the input.
 
 It saves the most on long-document retrieval, where it removes most of the
 input and keeps the evidence. On short coding-agent tool outputs it saves
 little by design. See [Which setting to use](#which-setting-to-use).
 
-trim-1 is the model inside [Caveman](https://github.com/JuliusBrussee/caveman)
-Engine, where the package is named `keep`.
+trim-1 is the model inside [Caveman](https://github.com/JuliusBrussee/caveman).
+The code is at [caveman-ai/trim-1](https://github.com/caveman-ai/trim-1).
 
 ## How it works
 
@@ -60,7 +60,7 @@ Three operating points ship:
 | `aggressive` | 0.035 | cut most |
 
 trim-1 fails closed. It returns the context unchanged when the query is blank,
-the context is under 2,000 characters, the context is over 8 MiB or 200,000
+the context is under 2,000 bytes, the context is over 8 MiB or 200,000
 lines, or the cut would remove less than 10% of the bytes.
 
 ## Results
@@ -200,20 +200,21 @@ values.
 
 ## Use
 
-trim-1 runs inside Caveman Engine. The feature code and the model are one Go
-package with no cgo, so it also builds for WebAssembly.
+The code is on GitHub: [caveman-ai/trim-1](https://github.com/caveman-ai/trim-1).
+It is pure Go with no dependencies, and the model is embedded.
 
-```go
-import "github.com/JuliusBrussee/caveman/engine/keep"
-
-res := keep.Compress(context, query, keep.Options{Point: "balanced"})
-fmt.Println(res.Output)
+```bash
+go install github.com/caveman-ai/trim-1/cmd/trim@latest
+trim --query "why does the build fail" build.log
 ```
 
-From the command line:
+From Go:
 
-```bash
-caveman-engine compress --query "why does the build fail" < build.log
+```go
+import trim "github.com/caveman-ai/trim-1"
+
+res := trim.Compress(context, query, trim.Options{Point: "balanced"})
+fmt.Print(res.Output)
 ```
 
 ## License