Model family

Qwen Models

Qwen models are strong choices for multilingual chat, coding, math, vision-language, and local developer workflows.

Alibaba QwenUpdated 2026ChatCodeReasoningVisionEmbeddingAgents

Best for

Chat

Use this family hub to compare Qwen variants for chat workflows, then open the detail page for deeper deployment notes.

Code

Use this family hub to compare Qwen variants for code workflows, then open the detail page for deeper deployment notes.

Reasoning

Use this family hub to compare Qwen variants for reasoning workflows, then open the detail page for deeper deployment notes.

Vision

Use this family hub to compare Qwen variants for vision workflows, then open the detail page for deeper deployment notes.

Variants

Qwen models grouped by workflow

Latest / flagship

ChatFrontier 2026reasoningcoding

Qwen3 235B A22B

Alibaba Qwen · Qwen

Best for: Builders testing frontier-style open-weight reasoning and coding in hosted or multi-GPU environments.

Details →
ChatFrontier 2026chatcoding

Qwen3.5

Alibaba Qwen · Qwen

Best for: Teams comparing current Qwen releases for chat, coding, multilingual, and agent workflows.

Details →
ChatFrontier 2026chatcoding

Qwen3.6

Alibaba Qwen · Qwen

Best for: Builders watching newer Qwen releases for assistant, coding, multilingual, and agent workflows.

Details →
ChatFrontier 2026chatmultilingual

Qwen3 235B A22B Thinking

Alibaba Qwen · Qwen

Best for: Multilingual chat, assistant workflows, and Qwen-family comparisons.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →
ChatFrontier 2026chatmultilingual

Qwen3 30B A3B

Alibaba Qwen · Qwen

Best for: Multilingual chat, assistant workflows, and Qwen-family comparisons.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →
ChatFrontier 2026chatmultilingual

Qwen3 32B

Alibaba Qwen · Qwen

Best for: Multilingual chat, assistant workflows, and Qwen-family comparisons.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →
ChatFrontier 2026chatmultilingual

Qwen3 14B

Alibaba Qwen · Qwen

Best for: Multilingual chat, assistant workflows, and Qwen-family comparisons.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →
ChatFrontier 2026chatmultilingual

Qwen3 8B

Alibaba Qwen · Qwen

Best for: Multilingual chat, assistant workflows, and Qwen-family comparisons.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →
CodeOpen weights where releasedcodingdeveloper

Qwen2.5 Coder 14B

Alibaba Qwen · Qwen

Best for: Coding assistants, repository help, and developer workflow evaluation.

Local: Use the exact checkpoint and quantization that matches your hardware and latency target.

Details →

Coding

Reasoning

Vision / multimodal

Embedding and reranking

Compare

All Qwen models in the directory

ModelTypeBest forLocal runner notesLicenseDetail
Qwen3 235B A22BChatBuilders testing frontier-style open-weight reasoning and coding in hosted or multi-GPU environments.Usually a server or multi-GPU model; use quantized builds or hosted inference for practical testing.Apache 2.0Open
Qwen3 VLVisionBuilders adding visual understanding to open AI workflows.Can be tested locally when compatible checkpoints and runtimes are available; multimodal serving is more demanding than text-only models.Check exact model cardOpen
Qwen3 CoderCodeDevelopers comparing open coding models for Continue, Aider, Cline, and Roo Code workflows.Smaller or quantized coder variants can be tested locally for IDE and coding-agent workflows.Check exact model cardOpen
Qwen3 EmbeddingEmbeddingBuilders who want a newer embedding family to compare against E5, BGE, and Jina.Smaller embedding variants are practical for local RAG and retrieval experiments.Check exact model cardOpen
Qwen3.5ChatTeams comparing current Qwen releases for chat, coding, multilingual, and agent workflows.Local fit depends on the exact checkpoint size and quantized builds.Check exact model cardOpen
Qwen3.6ChatBuilders watching newer Qwen releases for assistant, coding, multilingual, and agent workflows.Local fit depends on the exact checkpoint size, serving stack, and quantized builds.Check exact model cardOpen
Qwen3 235B A22B ThinkingChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen3 30B A3BChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen3 32BChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen3 14BChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen3 8BChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 72B InstructChatMultilingual chat, assistant workflows, and Qwen-family comparisons.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 Coder 32BCodeCoding assistants, repository help, and developer workflow evaluation.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 Coder 14BCodeCoding assistants, repository help, and developer workflow evaluation.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 Coder 7BCodeCoding assistants, repository help, and developer workflow evaluation.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 Math 72BReasoningMath-heavy prompts, reasoning tests, and educational workflow evaluation.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2.5 VLVisionVision-language assistants, document images, and multimodal agent workflows.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen
Qwen2 VLVisionVision-language baseline comparisons and multimodal prototypes.Use the exact checkpoint and quantization that matches your hardware and latency target.Check exact model cardOpen