# Qwen models on W&B Inference

## Qwen3 235B A22B Thinking-2507 inference overview

#### Price per 1M tokens

$0.10 (input)

$0.10 (output)

#### Parameters

22B (active)

235B (total)

#### Context window

262K

#### Release date

July 2025

### Qwen3 235B A22B Thinking-2507 inference details

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass. This “thinking-only” variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs in challenging domains. The model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases as of July 2025.

**Created by**: Alibaba  
**License**: apache-2.0  
**🤗 model card**: [Qwen3-235B-A22B-Thinking-2507](https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507)

```python

```

## Qwen3 235B A22B-2507 inference overview

#### Price per 1M tokens

$0.10 (input)

$0.10 (output)

#### Parameters

22B (active)

235B (total)

#### Context window

262K

#### Release date

July 2025

### Qwen3 235B A22B-2507 inference details

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, logical reasoning, math, code, and tool usage. The model does not implement “thinking mode” (<think> blocks). Compared to its base variant, this version delivers significant gains in knowledge coverage, long-context reasoning, coding benchmarks, and alignment with open-ended tasks. It is particularly strong on multilingual understanding, math reasoning (e.g., AIME, HMMT), and alignment evaluations like Arena-Hard and WritingBench.

**Created by**: Alibaba  
**License**: apache-2.0  
**🤗 model card**: [Qwen3-235B-A22B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507)

```python

```

## Qwen3 Coder 480B A35B inference overview

#### Price per 1M tokens

$1.00 (input)

$1.50 (output)

#### Parameters

35B (active)

480B (total)

#### Context window

262K

#### Release date

July 2025

### Qwen3 Coder 480B A35B  inference details

Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over repositories. The model features 480 billion total parameters, with 35 billion active per forward pass (8 out of 160 experts).

**Created by**: Alibaba  
**License**: apache-2.0  
**🤗 model card**: [Qwen3-Coder-480B-A35B-Instruct](https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct)

```python

```

## Qwen3 resources

### W&B Inference powered by CoreWeave guide

[Get started](https://wandb.ai/wandb_fc/product-announcements-fc/reports/Introducing-W-B-Inference-powered-by-CoreWeave--VmlldzoxMzI1NjU0MQ)

### Running Qwen3 Coder on W&B Inference

[Read the report](https://wandb.ai/wandb_fc/product-announcements-fc/reports/Run-Qwen3-Coder-on-W-B-Inference--VmlldzoxMzcxMjY1NA)

### AI engineering course: Agents

[Take the free course](https://wandb.ai/site/courses/agents/)