@llmtrim/cli 0.1.6 → 0.1.7

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +10 -7
  2. package/package.json +7 -7
package/README.md CHANGED
@@ -1,10 +1,13 @@
1
1
  # llmtrim
2
2
 
3
- **Cut ~66% off your LLM bill.** A drop-in local HTTPS proxy that compresses every LLM
4
- request and reply input, output, and cache with zero extra model calls. Works with
5
- Claude Code, Cursor, Cline, and any tool that talks to OpenAI / Anthropic / Google /
6
- DeepSeek / Mistral & co. Answers unchanged: every cut is re-counted with the provider's
7
- real tokenizer and auto-reverted if it doesn't save.
3
+ **llmtrim is a local proxy that compresses your LLM API requests so you pay less, with no
4
+ change to the answers.** It strips wasted tokens (verbose tool output, resent schemas,
5
+ bulky JSON, long context) out of every request before it reaches the provider: −31% input
6
+ and −74% output tokens, measured live across 112 A/B cases. Works with Claude Code, Cursor,
7
+ Cline, and any tool that talks to OpenAI / Anthropic / Google / DeepSeek / Mistral & co.
8
+
9
+ Every cut is re-counted with the provider's real tokenizer and auto-reverted if it doesn't
10
+ save, so it can never increase your bill or break a request.
8
11
 
9
12
  ```bash
10
13
  npm install -g @llmtrim/cli && llmtrim setup
@@ -13,11 +16,11 @@ llmtrim status --watch
13
16
  ```
14
17
 
15
18
  `setup` is transparent and fully reversible (`llmtrim uninstall`): a local CA, a proxy
16
- block in your shell profile, a background service. Everything runs locally nothing is
19
+ block in your shell profile, a background service. Everything runs locally; nothing is
17
20
  ever sent to us.
18
21
 
19
22
  This package installs a prebuilt native binary for your platform (Linux, macOS, Windows;
20
- x64 & arm64) no Rust toolchain needed.
23
+ x64 & arm64). No Rust toolchain needed.
21
24
 
22
25
  Docs, benchmarks (112 live A/B cases), and source: **https://github.com/fkiene/llmtrim**
23
26
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@llmtrim/cli",
3
- "version": "0.1.6",
3
+ "version": "0.1.7",
4
4
  "description": "Cut your LLM bill: drop-in proxy that compresses input, output, and cache. Any provider, answers unchanged.",
5
5
  "repository": "https://github.com/fkiene/llmtrim",
6
6
  "homepage": "https://github.com/fkiene/llmtrim#readme",
@@ -9,11 +9,11 @@
9
9
  "bin": { "llmtrim": "bin/llmtrim.js" },
10
10
  "files": ["bin", "README.md"],
11
11
  "optionalDependencies": {
12
- "@llmtrim/linux-x64": "0.1.6",
13
- "@llmtrim/linux-arm64": "0.1.6",
14
- "@llmtrim/darwin-x64": "0.1.6",
15
- "@llmtrim/darwin-arm64": "0.1.6",
16
- "@llmtrim/win32-x64": "0.1.6",
17
- "@llmtrim/win32-arm64": "0.1.6"
12
+ "@llmtrim/linux-x64": "0.1.7",
13
+ "@llmtrim/linux-arm64": "0.1.7",
14
+ "@llmtrim/darwin-x64": "0.1.7",
15
+ "@llmtrim/darwin-arm64": "0.1.7",
16
+ "@llmtrim/win32-x64": "0.1.7",
17
+ "@llmtrim/win32-arm64": "0.1.7"
18
18
  }
19
19
  }