Qwen

Summary pending.

• Qwen3.8 Max is now generally available as the successor to the Qwen3.8 Max Preview
• Supports 1,000,000 token context window for processing large inputs
• Pricing set at $2.00 per million input tokens and $6.00 per million output tokens

• Qwen3.7 Flash is now available as a vision-language reasoning model for multimodal tasks
• Supports 1,000,000 token context window
• Priced at $0.03 per million input tokens and $0.13 per million output tokens

• Qwen3.7-Plus model is now available with support for text and image input
• Context window expanded to 1,000,000 tokens
• Pricing set at $0.32 per million input tokens and $1.28 per million output tokens

• Qwen3.7-Max model released as flagship in the Qwen3.7 series
• Supports 1,000,000 token context window
• Optimized for agent-centric workloads with strengths in coding and productivity tasks
• Pricing set at $1.48 per million input tokens and $4.42 per million output tokens

Summary pending.

• Qwen3.6 Flash model is now available with support for text, image, and video inputs
• 1 million token context window enables processing of very long documents and conversations
• Tiered pricing structure introduced at $0.19 per million input tokens and $1.13 per million output tokens

• Qwen3.6-35B-A3B is a new open-weight multimodal model with 35 billion total parameters and 3 billion active parameters per token
• Uses a hybrid sparse mixture-of-experts architecture
• Supports a 262,144 token context window
• Priced at $0.14 per million input tokens and $1.00 per million output tokens

• Qwen3.6 27B is a new 27-billion-parameter language model with hybrid multimodal capabilities supporting text, image, and video inputs
• Offers a 262,144 token context window
• Priced at $0.30 per million input tokens and $2.00 per million output tokens

• Qwen3.5 Plus is now available as a large-scale multimodal model supporting text, image, and video inputs
• Model features a 1M token context window
• Pricing set at $0.30 per million input tokens and $1.80 per million output tokens

• Qwen 3.6 Plus now supports a 1,000,000 token context window
• Pricing has changed to $0.33 per million input tokens and $1.95 per million output tokens

• Qwen3.5-9B is a new 9-billion parameter multimodal model with reasoning, coding, and visual understanding capabilities
• Supports a 262,144 token context window
• Priced at $0.10 per million input tokens and $0.15 per million output tokens

• Qwen3.5-flash-02-23 uses a hybrid architecture combining linear attention with sparse mixture-of-experts for improved inference efficiency
• Supports 1,000,000 token context window
• Pricing set at $0.07 per million input tokens and $0.26 per million output tokens

• Qwen3.5 35B-A3B is a new vision-language model with hybrid architecture using linear attention and sparse mixture-of-experts
• Supports 262,144 token context window
• Pricing set at $0.14 per million input tokens and $1.00 per million output tokens

• Qwen3.5 27B is now available as a native vision-language model with linear attention mechanism
• Model supports 262,144 token context window
• Pricing set at $0.20 per million input tokens and $1.56 per million output tokens

• Qwen3.5 122B-A10B model released with hybrid architecture combining linear attention and sparse mixture-of-experts
• Supports 262,144 token context window
• Pricing set at $0.26 per million input tokens and $2.08 per million output tokens

• Qwen3.5 Plus models now use a hybrid architecture combining linear attention with sparse mixture-of-experts for improved inference efficiency
• Context window expanded to 1,000,000 tokens
• Pricing updated to $0.26 per million input tokens and $1.56 per million output tokens

• Qwen3.5 397B-A17B model released with hybrid architecture combining linear attention and sparse mixture-of-experts
• Supports 262,144 token context window
• Pricing set at $0.39 per million input tokens and $2.34 per million output tokens

• Qwen3-Max-Thinking model released as flagship reasoning model for complex multi-step cognitive tasks
• Context window of 262,144 tokens supports long documents and extended conversations
• Pricing set at $0.78 per million input tokens and $3.90 per million output tokens

• Qwen3-Coder-Next is now available as an open-weight model optimized for coding agents and local development
• Model uses sparse MoE design with 80B total parameters but only 3B activated per token
• Supports 262,144 token context window
• Priced at $0.12 per million input tokens and $0.80 per million output tokens

• Qwen3-VL-32B-Instruct is now available as a multimodal vision-language model supporting text, images, and video understanding
• Model supports a 131,072 token context window
• Pricing is $0.10 per million input tokens and $0.42 per million output tokens

• Qwen3-VL-8B-Thinking model released with reasoning optimization for visual and textual analysis
• Supports 131,072 token context window
• Pricing set at $0.18 per million input tokens and $2.10 per million output tokens

• Qwen3-VL-8B-Instruct is a new multimodal vision-language model supporting text, images, and video understanding
• Features improved multimodal fusion with Interleaved-MRoPE for long-horizon reasoning
• Offers a 262,144 token context window
• Priced at $0.12 per million input tokens and $0.45 per million output tokens

• Qwen3-VL-30B-A3B-Thinking model released with multimodal capabilities for text, images, and videos
• Thinking variant added to enhance reasoning for STEM, math, and complex tasks
• 262,144 token context window available
• Knowledge cutoff date is March 31, 2025

• Qwen3-VL-30B-A3B-Instruct is a new multimodal model supporting text generation and visual understanding for images and videos
• Model has a 262,144 token context window and knowledge cutoff of 2025-03-31
• Pricing is $0.13 per million input tokens and $0.52 per million output tokens

• Qwen3-VL-235B-A22B Thinking model released with multimodal capabilities for text and visual understanding
• Supports images and video with a 131,072 token context window
• Optimized for multimodal reasoning in STEM and math domains

• Qwen3-VL-235B-A22B Instruct is now available as an open-weight multimodal model supporting text generation and visual understanding for images and video
• Model supports a 262,144 token context window with knowledge cutoff at 2025-03-31
• Pricing set at $0.21 per million input tokens and $1.90 per million output tokens

• Improved reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version
• Context window increased to 262,144 tokens
• Knowledge cutoff updated to 2025-06-30
• Pricing adjusted to $0.78 per million input tokens and $3.90 per million output tokens

• Qwen3 Coder Plus is now available as a proprietary model variant
• Supports 1,000,000 token context window for handling large codebases
• Knowledge cutoff updated to June 30, 2025

• Qwen3 Coder Flash is now available as a fast, cost-efficient alternative to Qwen3 Coder Plus
• Supports 1 million token context window for handling large codebases
• Knowledge cutoff updated to June 30, 2025

• Qwen3-Next-80B-A3B-Thinking is a new reasoning-first chat model that outputs structured thinking traces by default
• Designed for complex multi-step problems including math proofs, code synthesis and debugging, logic, and agentic tasks
• Supports a 262,144 token context window with knowledge cutoff at 2025-09-30

• Qwen3-Next-80B-A3B-Instruct model released with 262,144 token context window
• Knowledge cutoff updated to 2025-09-30
• Optimized for fast, stable responses without thinking traces

• Qwen Plus model updated to 0728 version based on Qwen3 foundation with 1 million token context window
• Knowledge cutoff date is March 31, 2025
• Pricing adjusted to $0.26 per million input tokens and $0.78 per million output tokens

• Qwen3-30B-A3B-Thinking-2507 is a new 30B parameter Mixture-of-Experts reasoning model optimized for complex multi-step thinking tasks
• Model supports 81,920 token context window with knowledge cutoff at 2025-06-30
• Pricing set at $0.20 per million input tokens and $2.40 per million output tokens

• Qwen3-Coder-30B-A3B-Instruct is a new 30.5B parameter Mixture-of-Experts model with 128 experts designed for code generation and repository-scale understanding
• Supports a 262,144 token context window
• Knowledge cutoff updated to 2025-06-30
• Pricing set at $0.07 per million input tokens and $0.28 per million output tokens

• Qwen3-30B-A3B-Instruct model released with 30.5B parameters and 3.3B active parameters per inference
• Context window of 262,144 tokens supports long-form document processing
• Knowledge cutoff updated to 2025-06-30

• Qwen3-235B-A22B-Thinking-2507 is a new high-performance Mixture-of-Experts model optimized for complex reasoning tasks
• Supports a context window of 262,144 tokens
• Knowledge cutoff updated to 2025-06-30
• Pricing set at $0.23 per million input tokens and $2.30 per million output tokens

• Qwen3-Coder-480B-A35B-Instruct is a new Mixture-of-Experts code generation model optimized for agentic coding tasks
• Supports 262,144 token context window
• Knowledge cutoff date is June 30, 2025
• Pricing is $0.30 per million input tokens and $1.00 per million output tokens

• Qwen3-235B-A22B-Instruct-2507 is a new multilingual instruction-tuned mixture-of-experts model with 22B active parameters
• Supports a 262,144 token context window
• Knowledge cutoff updated to 2025-06-30
• Pricing set at $0.09 per million input tokens and $0.55 per million output tokens

• Qwen3-8B model is now available with 8.2B parameters supporting both reasoning and dialogue tasks
• Context window of 131,072 tokens enables processing of very long documents
• Knowledge cutoff updated to 2025-03-31