Sign In

@houtini/lm

Package Overview
Dependencies
Maintainers
1
Versions
33
Alerts
File Explorer

Advanced tools

License
Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

@houtini/lm - npm Package Compare versions

Comparing version
3.2.1
 to
3.2.2
+1
-1
package.json
{
"name": "@houtini/lm",
"version": "3.2.1",
"version": "3.2.2",
"type": "module",

@@ -5,0 +5,0 @@ "description": "MCP server for local LLMs — connects to LM Studio or any OpenAI-compatible endpoint",

+5
-2
README.md

@@ -44,2 +44,5 @@ # @houtini/lm Houtini LM - Save Tokens by Offloading Tasks from Claude Code to Your Local LLM Server (LM Studio / Ollama), Openrouter or a Cloud API

> New to local models? See **[docs/GETTING-STARTED.md](./docs/GETTING-STARTED.md)** — installing LM Studio or a Docker endpoint, getting an OpenAI-compatible URL for houtini, what the smaller models are good at, and which models fit on 16/32/64/96/128 GB of VRAM.
>
> Setting up a specific backend? Step-by-step guides, each with the traps that cause silent failures:
> **[LM Studio](./docs/SETUP-LMSTUDIO.md)** (easiest, desktop) · **[vLLM](./docs/SETUP-VLLM.md)** (throughput, tool-calling, long context).

@@ -469,6 +472,6 @@ ### Claude Code

|------|-----|-------|
| [LM Studio](https://lmstudio.ai) | `http://localhost:1234` | Default, zero config. Rich metadata via v0 API. |
| [LM Studio](https://lmstudio.ai) | `http://localhost:1234` | Default, zero config. Rich metadata via v0 API. **[Setup guide →](./docs/SETUP-LMSTUDIO.md)** |
| [Ollama](https://ollama.com) | `http://localhost:11434` | Set `HOUTINI_LM_ENDPOINT_URL`. Thinking models (qwen3, deepseek-r1) handled transparently — reasoning is captured from Ollama's `delta.reasoning` channel and the output budget is inflated automatically so small thinking models don't return empty bodies. |
| [OpenRouter](https://openrouter.ai) | `https://openrouter.ai/api` | 300+ models from one endpoint. Auto-detected — sends attribution headers, uses `reasoning.exclude` for thinking models, retries 429/5xx with jittered backoff, parallel requests allowed. |
| [vLLM](https://docs.vllm.ai) | `http://localhost:8000` | Native OpenAI API |
| [vLLM](https://docs.vllm.ai) | `http://localhost:8000` | Native OpenAI API. **[Setup guide →](./docs/SETUP-VLLM.md)** |
| [llama.cpp](https://github.com/ggml-org/llama.cpp) | `http://localhost:8080` | Server mode |

@@ -475,0 +478,0 @@ | [DeepSeek](https://platform.deepseek.com) | `https://api.deepseek.com` | 28c/M input tokens |

+2
-2
server.json

@@ -10,3 +10,3 @@ {

},
"version": "3.2.0",
"version": "3.2.2",
"packages": [

@@ -16,3 +16,3 @@ {

"identifier": "@houtini/lm",
"version": "3.2.0",
"version": "3.2.2",
"transport": [

@@ -19,0 +19,0 @@ {

dist/index.js

Sorry, the diff of this file is too big to display