Qwen: Qwen3 235B A22B Instruct 2507

Measured on Composite

This model has not carried enough traffic on Composite yet to publish a success rate, a latency figure or a traffic rank. The numbers appear here once it has.

Pricing

$0.109 per million tokens input · $0.438 per million tokens output

Credits pay the provider’s price plus 25%; cache hits cost less.

On a monthly plan, a typical call (8k tokens in, 1k out) uses about 1.1 plan credits, at the provider's own price.

Provider

cv11

Roleplay fit

Qwen models pair large context windows with solid multilingual ability, useful for roleplay that switches languages or references long lore documents.

Context behavior

With a 262,144-token context window, it can hold a long-running roleplay or a full lorebook in memory without losing earlier plot details, so it suits multi-session campaigns and character cards with heavy backstory.

Cost for regular use

It is priced low enough for high-volume daily roleplay without the per-message cost adding up quickly.

From the provider

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...

Supplied by cv11, not written by Composite.

Related models

Qwen: Qwen3.8 Max Prime · Qwen: Qwen3.8 Omni Flash · Qwen: Qwen3.8 Max (0902)