Lm vs Pqs MCP Server
A side-by-side look at the Lm and Pqs MCP Server MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Lm for conversational ai work as a local process; choose Pqs MCP Server for conversational ai as a local process. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Lm when
- You need dedicated capabilities in the Conversational AI domain.
- You prefer a local stdio subprocess transport.
- You have the keys it needs:
HOUTINI_LM_ENDPOINT_URL, HOUTINI_LM_API_KEY, HOUTINI_LM_MODEL, HOUTINI_LM_SERIALISE. - You want tools like
chat, custom_prompt, code_task.
Choose Pqs MCP Server when
- You need dedicated capabilities in the Conversational AI domain.
- You prefer a local stdio subprocess transport.
- You have the keys it needs:
PQS_BASE. - You want tools like
score_prompt, optimize_prompt.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Conversational AI | Conversational AI |
| Quality signal | 61/100 · Good | 56/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | Not declared |
| Pricing | Not declared | Not declared |
| Required env vars | HOUTINI_LM_ENDPOINT_URLHOUTINI_LM_API_KEYHOUTINI_LM_MODELHOUTINI_LM_SERIALISE | PQS_BASE |
| Verified owner | Community listing | Community listing |
| GitHub stars | 121 | 2 |
| Directory activity | 0 views 0 installs 0 upvotes | 2 views 0 installs 0 upvotes |
| Tools | 8 | 2 |
| Full listing | View Lm | View Pqs MCP Server |
Tools & capabilities
Lm tools 8
chatThe workhorse. Send a task, get an answer. The description includes planning triggers that nudge Claude to identify offloadable work when it's starting a big task.custom_promptThree-part prompt: system, context, instruction. Keeping them separate prevents context bleed - consistently outperforms stuffing everything into one message, especially with local models. I tested this properly one weekend - took the same batch of review tasks and ran them both ways. Splitting thi…code_taskBuilt for code analysis. Pre-configured system prompt with temperature and output constraints tuned per model family via the routing layer.code_task_filesLike `code_task`, but the local LLM reads files directly from disk — source never passes through the MCP client's context window. Use this when reviewing multiple related files, or a single large file that's awkward to paste. Files are read in parallel with `Promise.allSettled`, so one unreadable f…embedGenerate text embeddings via the OpenAI-compatible `/v1/embeddings` endpoint. Requires an embedding model to be available - Nomic Embed is a solid choice. Returns the vector, dimension count, and usage stats.discoverHealth check and speed readout. Returns model name, context window, capability profile, connection latency (labelled explicitly — this is the `/v1/models` fetch round-trip, *not* inference speed), and the active model's measured tok/s and TTFT averaged over the session. Before any real call has run…list_modelsLists everything on the LLM server - loaded and downloaded - with full metadata: architecture, quantisation, context window, capabilities, and HuggingFace enrichment data. Shows capability profiles describing what each model is best at, so Claude can make informed delegation decisions.statsCompact markdown dump of your offload stats — session and lifetime totals, per-model performance history, reasoning-token overhead — without the model catalog that `discover` prints. Cheap to call repeatedly to watch the 💰 counter climb.
Pqs MCP Server tools 2
score_promptScore a prompt's quality across 8 dimensions BEFORE sending it to an expensive model. Returns a 0-80 score, an A-F grade, the per-dimension breakdown (clarity, specificity, context, constraints, output_format, role_definition, examples, cot_structure), and the weakest dimension. USE WHEN: - The user is workshopping a prompt and asks "is this good?" / "will this work?" / "should I add more detail?" - The user is about to send a long or expensive prompt to GPT-4, Claude Opus, or any frontier model, especially in a batch or automation context where rework is costly. - The user mentions iterating on a prompt that produced poor output and wants to diagnose what's missing. - The user pastes a prompt and asks for feedback on it. DO NOT USE WHEN: - The user is asking you to write a prompt for them (write it yourself first, then optionally call score_prompt to verify). - The prompt is conversational chat (this scores task-shaped prompts). COST: Free, no API key required. Rate-limited per IP: 5/min, 10/day, 100/month. If the user exceeds the limit, the response will include a structured upgrade path with subscribe and account URLs. LATENCY: ~2 seconds.optimize_promptRewrite a prompt to score higher on the PQS rubric, AND show before/after output comparisons so the user can see the impact. Returns the optimized prompt, the original PQS score, the optimized PQS score, and side-by-side sample outputs from a frontier model using both versions. USE WHEN: - The user got a low score from score_prompt and asks how to improve. - The user explicitly asks to "improve" / "rewrite" / "fix" / "optimize" a prompt they pasted. - The user is dissatisfied with output quality from a previous prompt and asks how to get better results. - score_prompt returned a suggestion to invoke this tool. DO NOT USE WHEN: - The user just asked for a score (use score_prompt only — don't double up). - The user wants you to write a new prompt from scratch (write it directly). REQUIRES: A PQS API key from a Pro subscription ($19.99/month, 1,000 calls/mo, includes batch + A/B comparison). If the user has not provided one, the tool returns a clear subscription URL — pass that response to the user verbatim. Do not invent or guess API keys. There is no free trial of this tool; the user must subscribe before the first call. COST: Counted against your Pro subscription's monthly call quota. LATENCY: ~6-8 seconds.
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Lm
{
"mcpServers": {
"lm": {
"command": "npx",
"args": [
"-y",
"@houtini/lm"
],
"env": {
"HOUTINI_LM_ENDPOINT_URL": "YOUR_HOUTINI_LM_ENDPOINT_URL_HERE",
"HOUTINI_LM_API_KEY": "YOUR_HOUTINI_LM_API_KEY_HERE",
"HOUTINI_LM_MODEL": "YOUR_HOUTINI_LM_MODEL_HERE",
"HOUTINI_LM_SERIALISE": "YOUR_HOUTINI_LM_SERIALISE_HERE"
}
}
}
}
Pqs MCP Server
{
"mcpServers": {
"pqs-mcp-server": {
"command": "npx",
"args": [
"-y",
"pqs-mcp-server"
],
"env": {
"PQS_BASE": "YOUR_PQS_BASE_HERE"
}
}
}
}
Frequently asked questions
Lm is categorized under Conversational AI and uses a local stdio subprocess. Pqs MCP Server belongs to Conversational AI and uses a local stdio subprocess. Pick Lm for conversational ai work and Pqs MCP Server for conversational ai.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client.
Lm has no declared pricing — check the project's own site. Pqs MCP Server has no declared pricing — check the project's own site. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add lm and pqs-mcp-server as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Lm has a quality signal of 61/100 (Good) with 0 views, 0 installs and 121 GitHub stars. Pqs MCP Server has 56/100 (Good) with 2 views, 0 installs and 2 GitHub stars.