Requesty
Alibaba Cloud logo
Alibaba Cloud

qwen3-max

This is the best-performing model in the Qwen series. It is ideal for complex, multi-step tasks.

VisionTool callingCachingJSON schema
Try qwen3-maxCompareDocs

Specifications

Context window262K tokens
Max output66K tokens
API typechat
AddedSep 10, 2025
Model IDalibaba/qwen3-max
Data retentionNo
Used for trainingNo
Provider locationπŸ‡ΈπŸ‡¬ Singapore
Privacy policyAlibaba Cloud Privacy Policy→

Benchmarks

Released 2026-01-26
Coding Indexcoding
30.5%

Artificial Analysis Coding Index β€” a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
86.1%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
39.8%

Artificial Analysis Intelligence Index β€” a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality β€” always test on your own workload.

Pricing

Input / 1M
$0.86
Output / 1M
$3.44
Cache write
β€”
Cache read
β€”
Estimated cost
100K input + 10K output$0.12
1M input + 100K output$1.21
10M input + 1M output$12.05

Requesty charges exactly what the upstream provider charges β€” no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to alibaba/qwen3-max.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="alibaba/qwen3-max",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Alibaba Cloud models

qwen3.6-plus
1M ctx$0.50/1M
Coding 43
qwen3.7-max
1.0M ctx$2.50/1M
Coding 50
qwen3-30b-a3b-instruct-2507
131K ctx$0.20/1M
qwen3-coder-plus
1.0M ctx$1.00/1M
qwen3-coder-flash
1.0M ctx$0.30/1M
qwen-plus
131K ctx$0.40/1M

Frequently asked questions

How much does qwen3-max cost?
qwen3-max is priced at $0.86 per million input tokens and $3.44 per million output tokens when accessed via Requesty. Prompt caching is supported, which can cut effective input cost by up to 90% on repeated context. Requesty charges exactly what the upstream provider charges β€” we don't add markup.
What is the context window of qwen3-max?
qwen3-max has a context window of 262K tokens, with a maximum output of 66K tokens per response. That's roughly 350 words of input you can fit in a single prompt.
How does qwen3-max perform on benchmarks?
qwen3-max scores 86.1% on GPQA Diamond, 83.6% on τ²-Bench, 43.1% on SciCode. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can qwen3-max do?
qwen3-max supports vision input, tool calling, prompt caching, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use qwen3-max with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "alibaba/qwen3-max". The Quickstart above shows Python, JavaScript and cURL snippets.

Access qwen3-max through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Alibaba Cloud models