Qwen4 Max

qwen/qwen4-max
NewZDR
VisionToolsJSONReasoning
by Qwen · 2026-11-16

Qwen4 Max is the flagship large-scale mixture-of-experts (MoE) model from the Qwen team at Alibaba Cloud. It has a 1,000,000-token context window and accepts text, image and video input, returning text, with up to 65,536 output tokens in a single response. It supports reasoning / thinking with an adjustable thinking budget, built-in web search, native tool use / function calling, and structured (JSON) outputs, which makes it well suited to complex reasoning, multimodal understanding, long-context work and agentic pipelines. Served through an OpenAI-compatible Chat Completions API, the OpenAI Responses API, and the native Anthropic Messages API.

ctx1M tokens
Max output65K
Inputtext + image + video
Outputtext
p50 TTFT5.90 s
Data retentionZero Data Retention available. With ZDR on, requests go only to providers with a zero-retention agreement.
INPUT$1.50/ 1M tokens
OUTPUT$4.50/ 1M tokens
p50 TTFT5.90 s7d
p95 TTFT10.00 s7d
TRAFFIC2.0Mtokens / 7d

Qwen4 Max is the flagship large-scale mixture-of-experts (MoE) model from the Qwen team at Alibaba Cloud, with a 1,000,000-token context window. It accepts text, image and video input and returns…

What is Qwen4 Max?

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
  • Anthropic SDKhttps://api.orcarouter.ai
import os

from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key=os.environ["ORCAROUTER_API_KEY"],
)

response = client.chat.completions.create(
    model="qwen/qwen4-max",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • enable_search
  • enable_thinking
  • include_reasoning
  • logprobs
  • max_tokens
  • n
  • parallel_tool_calls
  • presence_penalty
  • reasoning
  • repetition_penalty
  • response_format
  • seed
  • stop
  • stream
  • stream_options
  • temperature
  • thinking_budget
  • tool_choice
  • tools
  • top_k
  • top_logprobs
  • top_p

Pricing

Pricing
Input / 1M tokens$1.50
Output / 1M tokens$4.50
Cache read / 1M$0.075
Cache write / 1M$1.875
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $24.00 · With prompt caching $19.01

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.002280

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
5.90 s
Output speed
1902 tok/s
p95 TTFT
10.00 s
Error rate
3.2%

Public benchmarks

Source: Design Arena

Community buzz

What developers are saying this week

Hacker News13 mentions · 7dup 6 vs the previous week

How it compares

Qwen4 MaxQwen3.8 MaxQwen3.8 Max (0902)qwen/qwen3-max-preview
Input $/M$1.50$2.00$2.00$0.86
Output $/M$4.50$6.00$6.00$3.44
Context1.0M1.0M1.0M262K
Quality8/109/109/108/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

What is Qwen4 Max?
The flagship large-scale mixture-of-experts (MoE) model from the Qwen team at Alibaba Cloud, with a 1,000,000-token context window. It accepts text, image and video input and returns text, with up to 65,536 output tokens.
What inputs does it accept?
Text, image and video in, text out, with reasoning / thinking, built-in web search, native tool use / function calling and structured (JSON) outputs.
How large is the context window?
Up to 1,000,000 tokens, with up to 65,536 output tokens in a single response.
How do I call it?
Use model qwen/qwen4-max with your OrcaRouter API key, over the OpenAI-compatible Chat Completions API, the OpenAI Responses API, or the native Anthropic Messages API.

Embed this badge

Qwen: Qwen4 Max•$1.50/M in•5898ms p50•via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/qwen/qwen4-max" target="_blank"> <img src="https://www.orcarouter.ai/embed/qwen/qwen4-max.svg" alt="Qwen: Qwen4 Max on OrcaRouter" /> </a>
Markdown [![Qwen: Qwen4 Max](https://www.orcarouter.ai/embed/qwen/qwen4-max.svg)](https://www.orcarouter.ai/models/qwen/qwen4-max)

Model card as data

GET /api/public/models/qwen/qwen4-maxOpen
Machine-readable:/llms.txt/llms-full.txt