Qwen3 235B (MoE)

Name: Qwen3 235B (MoE)
Author: Alibaba

AlibabaOpen WeightsPending Human Review

Qwen3-235B-A22B is the independently positioned flagship MoE tier of the Qwen3 family, with 235B total and 22B activated parameters. Its original April checkpoint combines thinking and non-thinking behavior in one model and supports a native 32K context, extendable through YaRN. The July 2507 updates offer separately trained Instruct and Thinking checkpoints with a native 256K context. This page represents the 235B tier specifically; the Qwen3 family page covers the broader size range. Apache-2.0 weights, named checkpoints, and hosted access support agent and reasoning applications.

2025-04-29

235B (22B active)

Mixture of Experts (MoE)

Apache 2.0

Compare with other models

Specifications

Parameters: 235B (22B active)
Architecture: Mixture of Experts (MoE)
License: Apache 2.0
Context Window: 32,768 tokens
Type: text
Modalities: text

Benchmark Scores

Advanced Specifications

Model Family: Qwen
API Access: Available
Chat Interface: Available
Multilingual Support: Yes
Variants: Qwen3-235B-A22BQwen3-235B-A22B-Instruct-2507Qwen3-235B-A22B-Thinking-2507
Hardware Support: CUDATPU

Capabilities & Limitations

Capabilities: hybrid reasoningthinking modenon-thinking modemathematicscodinglogical reasoningfunction-callingtool usecreative writingmultilingual
Known Limitations: Context specifications differ between April and July checkpoints.Full 235B expert weights require substantial serving memory.
Notable Use Cases: agent-based applicationscomplex reasoning tasksmultilingual applicationsAlibaba's Quark AI assistantComplex reasoningScientific researchAgentic workflows
Function Calling Support: Yes
Tool Use Support: Yes

Resources

Related Models

IndustryLLM

Alibaba

IndustryLLM is Alibaba’s open-weight language model for industrial procurement and engineering-standard reasoning. It adapts Qwen3.5-35B-A3B-Base through continued pre-training on industrial documents and transactions followed by task-focused supervised fine-tuning. The 35.95B-parameter hybrid mixture-of-experts network activates approximately 3B per token. Training targets informal buyer language, ambiguous product codes, contradictory specifications, and evidence-based constraint evaluation. Apache-licensed BF16 weights are published. Its 262K architectural context is distinct from 12,288-token packed fine-tuning sequences; this record uses the supported training length conservatively. The September 25 paper predates later discovery-index coverage.

Typetext

Parameters35.95B total, approximately 3B active

2026-09-25

Open Weights

Details Compare

Qwen-Audio-3.1-Realtime

Alibaba

Qwen-Audio-3.1-Realtime is a proprietary language-capable speech conversation model for fluid real-time dialogue. It combines speech reasoning with duplex interaction, allowing conversational responses rather than only transcription or speech synthesis. The 3.1 Plus update supports a 262,144-token context, function calling, web search, voice cloning, and additional system voices, while retaining the 3.0 Plus integration protocol. Hosted streaming APIs provide access; no downloadable weights or checkpoint parameter count is verified.

Qwen3.8-Omni-Flash

Alibaba

Qwen3.8-Omni-Flash is a native omnimodal agent model built on the Qwen3.8-Flash-Next architecture. It accepts text, images, audio, and video in a million-token context, extending coding, GUI operation, and knowledge work to audio-visual productivity. Standard APIs return text, while the subsequently available Realtime variant supports text and speech output, multichannel audio, video aggregation, and remote MCP tools. Hosted access and companion plugins support video editing, multimedia summarization, and real-time conversations; its weights are not publicly released.