Qwen3 Max

Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version. It delivers higher accuracy in math, coding, logic, and science tasks, follows complex instructions in Chinese and English more reliably, reduces hallucinations, and produces higher-quality responses for open-ended Q&A, writing, and conversation. The model supports over 100 languages with stronger translation and commonsense reasoning, and is optimized for retrieval-augmented generation (RAG) and tool calling, though it does not include a dedicated “thinking” mode.

alibaba/qwen3-max

Organization: Alibaba
Family: qwen
Providers: 15
Context: 262,144
Output limit: 65,536
Knowledge: 2025-04
Release: 2025-09-23
Updated: 2025-09-23
Weights: Closed
Input: text
Output: text
Capabilities: Tools, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

18 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
AIHubMix	qwen3-max	262,144	65,536	$0.4508 / $2.7048	No	Yes	—	View
Cloudflare AI Gateway	alibaba/qwen3-max	262,144	65,536	$1.20 / $6.00	No	Yes	—	View
Deep Infra	Qwen/Qwen3-Max	256,000	65,536	$1.20 / $6.00	No	Yes	Yes	View
DevPass (LLM Gateway)	qwen3-max	262,144	65,536	$0.845 / $3.38	No	Yes	—	View
Eden AI	qwen/qwen3-max@eu	262,144	65,536	$1.20 / $6.00	No	Yes	Yes	View
Eden AI	qwen/qwen3-max	262,144	65,536	$1.20 / $6.00	No	Yes	Yes	View
EmpirioLabs AI	qwen3-max	256,000	65,536	$1.08 / $5.52	Yes	Yes	Yes	View
Kilo Gateway	qwen/qwen3-max	262,144	65,536	$0.78 / $3.90	No	Yes	Yes	View
LLM Gateway	novita/qwen3-max	262,144	65,536	$0.845 / $3.38	No	Yes	No	View
LLM Gateway	alibaba/qwen3-max	256,000	32,800	$1.20 / $6.00	Yes	Yes	No	View
LLMTR	qwen/qwen3-max	256,000	65,536	$1.20 / $6.00	Yes	Yes	—	View
Merge Gateway	qwen/qwen3-max	262,144	65,536	$0.359 / $1.434	No	Yes	Yes	View
NanoGPT	qwen/qwen3-max	256,000	32,768	$1.2002 / $6.001	No	No	No	View
Ofox	qwen/qwen3-max	256,000	64,000	$0.36 / $1.43	Yes	Yes	—	View
Ofox	bailian/qwen3-max	262,144	65,536	$0.36 / $1.43	Yes	Yes	—	View
Alibabavia OpenRouter	qwen/qwen3-max unknown	262,144	65,536	$0.78 / $3.90	No	Yes	Yes	View
OrcaRouter	qwen/qwen3-max	262,144	65,536	$0.359 / $1.434	No	Yes	—	View
Vercel AI Gateway	alibaba/qwen3-max	262,144	32,768	$1.20 / $6.00	No	Yes	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.359Per 1M input tokens

Output from$1.43Per 1M output tokens

Available offers18Across 15 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 1 lowest-priced configuration

InputOutput

Alibabaalibaba

$0.78$3.90

qwen/qwen3-maxObserved Oct 4, 2026, 11:31 AM UTC

Input: $0.78
Output: $3.90
Cache read: $0.156
Cache write: $0.975

Additional fees and pricing conditions

overrides · USD per token for token prices; time windows use UTC.

[
  {
    "prompt": "0.00000156",
    "completion": "0.0000078",
    "input_cache_read": "0.000000312",
    "input_cache_write": "0.00000195",
    "min_prompt_tokens": 32000
  },
  {
    "prompt": "0.00000195",
    "completion": "0.00000975",
    "input_cache_read": "0.00000039",
    "input_cache_write": "0.0000024375",
    "min_prompt_tokens": 128000
  }
]

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
Alibabaalibaba	$0.78	$3.90	$0.156	$0.975

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 1 endpoint

P50P75P90P99

Alibabaalibaba

767.5–2,339.25 ms

All latency percentiles 1 endpoints

Endpoint	P50	P75	P90	P99
Alibabaalibaba	767.5 ms	982 ms	1,374.2 ms	2,339.25 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 1 endpoint

P50P75P90P99

Alibabaalibaba

55–74.84 t/s

All throughput percentiles 1 endpoints

Endpoint	P50	P75	P90	P99
Alibabaalibaba	55 t/s	62 t/s	67 t/s	74.84 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 1 endpoint

Endpoint5 minutes30 minutes24 hours

AlibabaalibabaObserved Oct 4, 2026, 11:31 AM UTC

100%

99.8%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

qwen/qwen3-maxObserved Oct 4, 2026, 11:31 AM UTC

Design Arena	Category	Elo	Win rate	Rank
models	3d	1,089	43.5%	89
models	asciiart	1,139	47.2%	50
models	codecategories	1,116	44%	90
models	dataviz	1,109	41.2%	90
models	gamedev	1,107	43.8%	87
models	svg	1,010	37.2%	82
models	uicomponent	1,085	40.2%	94
models	website	1,125	44.5%	92

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation	Configuration	Source	Results	Details
models · uicomponent	Qwen3 Maxqwen/qwen3-max	Design Arena	1,085 Elo	Details
models · asciiart	Qwen3 Maxqwen/qwen3-max	Design Arena	1,139 Elo	Details
models · 3d	Qwen3 Maxqwen/qwen3-max	Design Arena	1,089 Elo	Details
models · website	Qwen3 Maxqwen/qwen3-max	Design Arena	1,125 Elo	Details
models · svg	Qwen3 Maxqwen/qwen3-max	Design Arena	1,010 Elo	Details
models · gamedev	Qwen3 Maxqwen/qwen3-max	Design Arena	1,107 Elo	Details
models · codecategories	Qwen3 Maxqwen/qwen3-max	Design Arena	1,116 Elo	Details
models · dataviz	Qwen3 Maxqwen/qwen3-max	Design Arena	1,109 Elo	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

qwen/qwen3-maxObserved Oct 4, 2026, 11:31 AM UTC

Context: 262,144
Maximum output: 65,536
Tokenizer: Qwen3
Input: text
Output: text
Moderation: No
Knowledge cutoff: 2025-06-30
Retirement: 2026-10-09
Added to OpenRouter: 2025-09-23

Base URLhttps://openrouter.ai/api/v1

Model IDqwen/qwen3-max

Version IDqwen/qwen3-max

Reasoning

Required: No
Enabled by default: Not reported
Default effort: —
Supported efforts: —

Supported parameters

frequency_penalty
logprobs
max_tokens
presence_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Default parameters

top p: 1
temperature: 1

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Qwen3 Max cost?

Listed rates start at $0.359 per million input tokens and $1.43 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Qwen3 Max?

There are 18 cataloged offers across 15 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 262,144 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Sep 14, 2026 · cbcb1fd10668
Sep 12, 2026 · 392c9aed8402