Qwen3.8-Max

Alibaba Cloud released Qwen3.8-Max-0902, an upgraded snapshot of its Qwen3.8-Max large language model, featuring enhanced coding capabilities for complex engineering projects, improved collaborative agent performance, and refined native vision understanding across chart reasoning and document parsing. The model retains a 1M context window, thinking mode, and full tool ecosystem, with pricing set at $2 per 1M input tokens and $6 per 1M output tokens.

Qwen3.8-Max Copy success Try AI https://www.qwencloud.com/try-ai/chat?models=qwen3.8-max-0902 Add to Compare /compare?models=qwen3.8-max-0902 Overview Qwen3.8-Max-0902（alias qwen3.8-max-2026-09-02）is an upgraded snapshot of qwen3.8-max. Coding capability breaks new ground, handling more complex engineering-scale projects and long-horizon autonomous development. Collaborative agent performance is significantly enhanced, with greater composure in multi-tool orchestration and end-to-end task delivery. Native vision understanding is refined across chart reasoning, document parsing, and multimodal perception — sharper and more reliable. Retains the 1M context window, thinking mode, and full tool ecosystem, evolving at a higher level of intelligence. Input ImageTextVideo Output Text Features Prefix Completion Enable Partial Mode when calling the Qwen API to make the model continue strictly from your provided prefix text. View docs https://docs.qwencloud.com/developer-guides/text-generation/partial-mode Function Calling Use function calling to connect large language models with external tools and systems. View docs https://docs.qwencloud.com/developer-guides/text-generation/function-calling Cache Context Cache stores shared prefixes for long-context requests to reduce repeated computation, improve latency, and lower cost. View docs https://docs.qwencloud.com/developer-guides/text-generation/context-cache implicit-cache Structured Outputs Structured Outputs help ensure the model returns a JSON string in the expected format. View docs https://docs.qwencloud.com/developer-guides/text-generation/structured-output Pricing - Input$2Per 1M tokens - Output$6Per 1M tokens - Input Implicit Cache $0.25Per 1M tokens - Explicit Cache Creation$2.5Per 1M tokens - Explicit Cache Read$0.17Per 1M tokens Rate Limits & Context - Max Input991K - Max Output131K - Max Input Thinking 983K - Max Output Thinking 131K - Context1M - Max Reasoning262K - TPMTokens Per Minute1M - RPMRequests Per Minute15K Built-in Tools code interpreter https://docs.qwencloud.com/developer-guides/text-generation/code-interpreter Responses API i2i search https://docs.qwencloud.com/developer-guides/text-generation/image-search Responses API t2i search https://docs.qwencloud.com/developer-guides/text-generation/image-search Responses API web extractor https://docs.qwencloud.com/developer-guides/text-generation/web-scraping Responses API web search https://docs.qwencloud.com/developer-guides/text-generation/web-search Responses API API Reference Call API https://home.qwencloud.com/api-keys Copy success 123456789101112131415161718192021222324252627282930 python from openai import OpenAI import os client = OpenAI If the environment variable is not set, replace it with your Model Studio API key: api key="sk-xxx" api key=os.getenv "DASHSCOPE API KEY" , base url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1", messages = {"role": "user", "content": "Who are you"} completion = client.chat.completions.create model="qwen3.8-max-0902", You can replace this with another deep thinking models messages=messages, extra body={"enable thinking": True}, stream=True is answering = False Indicates whether the response phase has started print "\n" + "=" 20 + "Thinking process" + "=" 20 for chunk in completion: if not chunk.choices: continue delta = chunk.choices 0 .delta if hasattr delta, "reasoning content" and delta.reasoning content is not None: if not is answering: print delta.reasoning content, end="", flush=True if hasattr delta, "content" and delta.content: if not is answering: print "\n" + "=" 20 + "Full response" + "=" 20 is answering = True print delta.content, end="", flush=True 123456789101112131415161718192021222324252627282930 python from openai import OpenAI import os client = OpenAI If the environment variable is not set, replace it with your Model Studio API key: api key="sk-xxx" api key=os.getenv "DASHSCOPE API KEY" , base url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1", messages = {"role": "user", "content": "Who are you"} completion = client.chat.completions.create model="qwen3.8-max-0902", You can replace this with another deep thinking models messages=messages, extra body={"enable thinking": True}, stream=True is answering = False Indicates whether the response phase has started print "\n" + "=" 20 + "Thinking process" + "=" 20 for chunk in completion: if not chunk.choices: continue delta = chunk.choices 0 .delta if hasattr delta, "reasoning content" and delta.reasoning content is not None: if not is answering: print delta.reasoning content, end="", flush=True if hasattr delta, "content" and delta.content: if not is answering: print "\n" + "=" 20 + "Full response" + "=" 20 is answering = True print delta.content, end="", flush=True