# Qwen 3.6 27B

> Canonical: https://www.overmindlab.ai/models/qwen-3-6-27b

Qwen mid-tier base for coding and maths-heavy agents you plan to own.

| Property | Value |
| --- | --- |
| Model ID | Qwen/Qwen3.6-27B |
| Provider | Qwen |
| Group | Qwen 3.6 |
| Parameters | 27B |
| Tier | Mid |
| Context window | 256K |
| Max training context | 256K |
| Tool calling | Yes |
| Training methods | — |
| Training cost | from $2.00 |
| Serving cost (per 1M output tokens) | from $15.00 |

## Overview

Qwen 3.6 is a mid-tier Qwen generation Overmind trains. Use it when you want published coding and maths scores in the Qwen line as a mid-tier base, then train on your agent's own traces and benchmark against the model you already ship.

Qwen 3.6 on the platform. Mid tier, 256K context window.

## Good for

- Coding and maths-heavy agent loops
- Mid-tier quality before committing to a larger serve cost
- Parallel experiments against a compact Qwen 3.5 baseline

Train this model on your agent's data: https://docs.overmindlab.ai/models/training.md