# Qwen 3.5 35B MoE

> Canonical: https://www.overmindlab.ai/models/qwen-3-5-35b-a3b

Large MoE Qwen 3.5: 35B total weights, 3B active per token at serve time.

| Property | Value |
| --- | --- |
| Model ID | Qwen/Qwen3.5-35B-A3B |
| Provider | Qwen |
| Group | Qwen 3.5 |
| Parameters | 35B MoE (3B active) |
| Tier | Large |
| Context window | 256K |
| Max training context | 256K |
| Tool calling | Yes |
| Training methods | — |
| Training cost | from $2.50 |
| Serving cost (per 1M output tokens) | from $8.00 |

## Overview

Qwen 3.5 is the default recommendation for most agents on Overmind. Every size shares a 256K inference context, tool calling starts at 0.8B, and the compact models train at their full 256K context, the longest trainable window in the catalogue.

The default recommendation for most agents. A 256K context window at every size, tool calling from 0.8B up, and compact models trainable at their full 256K context, the longest in the catalogue.

## Good for

- Higher-quality Qwen 3.5 runs with MoE serve economics
- Tool-calling agents that outgrow dense mid-tier bases
- Side-by-side experiments against Qwen 3.5 27B

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

## Related models

- [Qwen 3.5 4B](https://www.overmindlab.ai/models/qwen-3-5-4b): 4B, 256K context
- [Qwen 3.5 2B](https://www.overmindlab.ai/models/qwen-3-5-2b): 2B, 256K context
- [Qwen 3.5 0.8B](https://www.overmindlab.ai/models/qwen-3-5-0-8b): 0.8B, 256K context
- [Qwen 3.5 9B](https://www.overmindlab.ai/models/qwen-3-5-9b): 9B, 256K context