Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d268Policy30d67, 15 pendingSources1,014

Qwen3.8 Max Preview

AlibabaReleased Aug 3, 2026Current

Score basis

Qwen3.8 Max Preview is at the frontier in long context; strong in agentic tasks; capable in instruction following, factuality, reasoning, coding, and multimodal tasks; and behind the leaders in math. Too few results yet to rate safety or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Frontier1 capability
  1. Long Context−1.8%2 of 3
Strong1 capability
  1. Agentic−9.2%5 of 7
Capable5 capabilities
  1. Instruction Following−10.8%2 of 3
  2. Factuality−14.0%3 of 4
  3. Reasoning−15.4%4 of 6
  4. Coding−15.6%6 of 10
  5. Multimodal−23.1%1 of 6
Limited1 capability
  1. Math−26.0%5 of 5
Not rated2 capabilities

Too few results yet to rate Safety or Multilingual.

Price

$2.00input$6.00outputper million tokens

From Artificial Analysis · 4 providers tracked · All prices

Costs more than 75% of 331 priced models · 3:1 input-to-output blend, log scale

Evidence

87results on59benchmarks

  • 19 independently verified
  • 32 aggregator
  • 27 vendor-reported
  • 9 cross-referenced

From 12 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

7 papers reference Qwen3.8 Max Preview

Qwen3.8 Max Preview benchmark results

87 results on 59 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Long Context

FrontierFull long context ranking

1.8% behind the leader2 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Agentic

StrongFull agentic ranking

9.2% behind the leader5 of 7 ranked benchmarks measured

Show 5 more agentic resultsHide 5 agentic results

Instruction Following

CapableFull instruction following ranking

10.8% behind the leader2 of 3 ranked benchmarks measured

Factuality

CapableFull factuality ranking

14.0% behind the leader3 of 4 ranked benchmarks measured

Show 3 more factuality resultsHide 3 factuality results

Reasoning

CapableFull reasoning ranking

15.4% behind the leader4 of 6 ranked benchmarks measured

Show 5 more reasoning resultsHide 5 reasoning results

Coding

CapableFull coding ranking

15.6% behind the leader6 of 10 ranked benchmarks measured

Show 5 more coding resultsHide 5 coding results

Multimodal

CapableFull multimodal ranking

23.1% behind the leader1 of 6 ranked benchmarks measured

Show 3 more multimodal resultsHide 3 multimodal results

Math

LimitedFull math ranking

26.0% behind the leader5 of 5 ranked benchmarks measured

Show 2 more math resultsHide 2 math results

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 32 more resultsHide 32 results

Qwen3.8 Max Preview: common questions

Who makes Qwen3.8 Max Preview?

Qwen3.8 Max Preview is made by Alibaba.

When was Qwen3.8 Max Preview released?

Qwen3.8 Max Preview was released on Aug 3, 2026, according to Artificial Analysis.

What is Qwen3.8 Max Preview good at?

Qwen3.8 Max Preview is at the frontier in long context; strong in agentic tasks; capable in instruction following, factuality, reasoning, coding, and multimodal tasks; and behind the leaders in math. Too few results yet to rate safety or multilingual tasks.

How much does Qwen3.8 Max Preview cost?

Qwen3.8 Max Preview costs $2.00 per million input tokens and $6.00 per million output tokens, according to Artificial Analysis. We track its price at 4 providers. At a mix of three input tokens to one output token, it costs more than 75% of the 331 priced models we track.

How many benchmarks has Qwen3.8 Max Preview been tested on?

We track 87 results for Qwen3.8 Max Preview on 59 benchmarks from 12 sources, 19 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does Qwen3.8 Max Preview support?

OpenRouter lists tool calling, structured outputs, and reasoning for Qwen3.8 Max Preview.

About this record

Where Qwen3.8 Max Preview's numbers come from, and every name it appears under.

Tracked since
Jul 19, 2026
Newest source mention
Sep 22, 2026

Where the results come from

Verification: 87 scores · 19 independently verified · 32 aggregator-attributed · 9 vendor cross-reference · 27 vendor-reported. How these tiers are assigned

From 12 sources on 7 sites. Artificial Analysis supplies 34 of them; the 19 independently verified results come from 5 sites. Bars are coloured by trust tier.

  • artificialanalysis.ai34
  • api.llm-stats.com23
  • huggingface.co13
  • livebench.ai7
  • epoch.ai6
  • datasets-server.huggingface.co3
  • lmarena.ai1

Also known as

quent 3.8 maxqwen 3.8 maxqwen3-8-maxqwen3.8 maxqwen3.8 max (xhigh)Qwen3.8-Max-0902qwen 3.8-max-0902qwen3.8 max (0902) (xhigh)qwen3.8 max (0902)qwen3-8-max-0803qwen3.8-max-prime