# Qwen 3.8 27B

> Canonical: https://www.overmindlab.ai/models/qwen-3-8-27b

Newest Qwen mid-tier base for coding and long-horizon agents you plan to own.

| Property | Value |
| --- | --- |
| Model ID | Qwen/Qwen3.8-27B |
| Provider | Qwen |
| Group | Qwen 3.8 |
| 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 3.8 is the newest Qwen generation Overmind trains. Use it when you want the current Qwen coding and agent scores as a mid-tier base, then train LoRA on your agent's own traces.

The newest Qwen generation on the platform. Dense 27B vision-language model, 256K native context, LoRA-only training.

## Good for

- Coding and long-horizon agent loops
- Mid-tier quality before committing to a larger serve cost
- Parallel experiments against Qwen 3.6 and 3.5 27B

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