Macaron V1 Tall is the smaller of Mind Lab's two Macaron V1 releases, pairing a 35 billion parameter base post-trained from Qwen 3.6 with four 3.7 billion parameter LoRA specialists for a 50 billion parameter total. It keeps the Mixture-of-LoRA design of the larger Venti build, where a router picks a specialist per request rather than running one set of weights for every kind of task. The size is the point. Tall is meant to run locally, on hardware that could never host the 748 billion parameter Venti, while keeping the same per-task specialization and continual-learning approach. That fits personal agents, coding assistance and terminal work on self-managed infrastructure, where keeping the model close to the data matters more than the last increment of capability.
Macaron V1 Tall runs on 1 route through the qynio gateway. Compare residency, zero retention, and training posture at a glance — full data-handling detail per route below.
| Provider | Region | Zero data retention | Training | Input | Output |
|---|---|---|---|---|---|
| US | Zero data retention | No | $0.45 | $2.60 |
Macaron V1 Tall runs on a single route through the qynio gateway today, so its availability is that provider's availability.
Name a second model in the same request and the gateway tries it on retriable errors, so a busy hour never has to reach your users. .
Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. See uptime for every provider qynio monitors.
Each route hosting Macaron V1 Tall has its own privacy posture, residency, and no-KYC terms. Postures are maintained by qynio with a last-verification timestamp.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. US; wallets; deposit available.
Call Macaron V1 Tall through the qynio gateway with one API key. Let your coding agent set it up, or call it directly — qynio is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.
Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up qynio for you.
import OpenAI from "openai";const client = new OpenAI({apiKey: process.env.QYN_API_KEY,baseURL: "https://api.qyn.io/v3/compat",});const completion = await client.chat.completions.create({model: "novita/mindai/macaron-v1-tall",messages: [{ role: "user", content: "Hello" }],});console.log(completion.choices[0].message.content);
Side-by-side on privacy, crypto rails, pricing, and benchmarks.
