Is Qwen 3.7 Max available with Zero Data Retention on Elosia?

Qwen 3.7 Max runs on Elosia's standard backend; ZDR is available on many other models in the catalog.

How much does Qwen 3.7 Max cost on Elosia?

Qwen 3.7 Max is billed by usage at $1.475 per 1M input tokens and $4.425 per 1M output tokens, within your Elosia plan.

What is Qwen 3.7 Max's context window?

Qwen 3.7 Max supports a context window of up to 1,000,000 tokens.

Is Qwen 3.7 Max hosted in the EU?

Qwen 3.7 Max is not available on Elosia's EU-hosted (EdenAI) backend; it runs on the default backend.

When should you choose Qwen 3.7 Max?

Qwen 3.7 Max is a balanced model: a strong quality/cost trade-off for everyday production work.

Qwen 3.7 Max

QwenBalanced

ThinkingTool UseStructured Output

About this model

Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...

Performance Tier

Balanced

Qwen 3.7 Max is a balanced model from Qwen : strong performance at a reasonable price.

Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.

Pricing

This model is included in Elosia plans

Moderate

Moderate cost. A balanced choice for regular use.

Type	per 1M tokens
Input (prompt)	$1.48
Output (completion)	$4.42
Cache read	$0.295
Cache write	$1.84

Capabilities

Context Length1.0M

Max Output Tokens131K

TokenizerQwen

Inputtext

Outputtext

Release DateMay 21, 2026

Benchmarks

General Intelligence

MMLU

Not reported

GPQA Diamond

92.4%

Mathematics

MATH-500

Not reported

AIME 2025

93.3%

Programming

HumanEval

Not reported

SWE-bench Verified

80.4%

SWE-bench Multilingual

78.3%

Reasoning

Humanity's Last Exam

41.4%

Agentic

SWE-bench Pro

60.6%

Terminal-Bench 2.0

69.7%

Where does Qwen 3.7 Max stand?

Compare its performance index against every other model.

View the performance leaderboard

Recommended Use Cases

CodingAnalysisResearchGeneral ChatTranslation

Strengths

Agent model of the Qwen line, evaluated on real repository patches rather than isolated function completion
Scientific reasoning and competition mathematics are part of the reported suite, so the scope is not code alone
Agentic coding evaluated across several programming languages, not on a single-language repository set
Long-horizon command-line workflows covered as well, with terminal agentic tasks reported next to repository patches
Built-in thinking mode (enable_thinking / preserve_thinking) for chain-of-thought reasoning

Limitations

Alibaba's evaluation focuses on agentic and coding benchmarks: MMLU, MATH-500, HumanEval and LiveCodeBench are not part of the reported suite
Text input only, with no image or video accepted, and no parallel tool calls exposed
Closed weights with no published checkpoint, so no self-hosted deployment