Qwen3-Coder

Name: Qwen3-Coder
Author: Alibaba

AlibabaOpen WeightsPending Human Review

Qwen3-Coder is an open-weight MoE coding family for agentic software development. Its flagship has 480B total parameters and 35B activated, with native 256K context extendable through YaRN. Training targets repository-scale understanding, code editing, execution feedback, and function calling, with the Qwen Code agent providing an integration example. A smaller 30B-A3B checkpoint is grouped as a routine size variant. The instruct models are non-thinking and use a dedicated tool-call format; Apache-2.0 weights enable local serving.

2025-07-22

480B total, 35B active (flagship); 30B total, 3B active (smaller tier)

Mixture of Experts (MoE) Transformer

Apache-2.0

Compare with other models

Specifications

Parameters: 480B total, 35B active (flagship); 30B total, 3B active (smaller tier)
Architecture: Mixture of Experts (MoE) Transformer
License: Apache-2.0
Context Window: 262,144 tokens
Type: text
Modalities: text

Benchmark Scores

Advanced Specifications

Model Family: Qwen
API Access: Available
Chat Interface: Not Available
Variants: Qwen3-Coder-480B-A35B-InstructQwen3-Coder-30B-A3B-InstructQwen3-Coder-Plus (hosted flagship)Qwen3-Coder-Flash (hosted smaller tier)

Capabilities & Limitations

Capabilities: code generationrepository understandingfunction callingagentic coding
Known Limitations: Self-hosting requires sufficient memory for all model weights, including inactive experts.Reasoning and agent benchmark results depend on the harness and sampling settings.The instruct checkpoints operate in non-thinking mode.Million-token operation requires context-extension configuration rather than the native default.
Notable Use Cases: software engineering agentsrepository maintenanceIDE assistants
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.