Skip to content

Alibaba · Chat

Qwen3.7 Max

Alibaba 🇨🇳ChatQwen familyOne of its sources says the weights aren't publicly available.

Newer version: Qwen4 Max. We compared release dates and version names in our catalogue.

From Empiriolabs's listing:
Qwen3.7 Max is a flagship text model for coding, productivity, long-running agents, deep thinking, tools, and 1M-token context.
llminfo IQ
134.9
Maker's price
$3.75 blended / 1M
Max context
1M
Listings
55

Evidence and sources

International storefront (alibabacloud.com) is Singapore-incorporated; mainland operating entity 杭州阿里云智能科技有限公司 (Hangzhou).

llminfo IQ 134.9, ranked among chat models. From 225 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 112.2, 95% interval 107–117. From 24 benchmark families across 9 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedPoor value (44.2). Costs more per task than its benchmark results predict. From 24 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsTypical (55 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.Median 50,182 output tokens per task at default effort.How the Yap score worksReleased 19 May 2026Maker HQ: CN (source)
All rankings: #232 on OpenRouter this month #233 on OpenRouter this week #259 on OpenRouter today #569 on NanoGPT recentlySources: LMArena data, DeepInfra data, EmpirioLabs data, Models.dev data, llminfo researchPrimary data from LMArena from DeepInfra: name from EmpirioLabs: primary mode, description from Models.dev: weights availability from llminfo research: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 10 maker's prices.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb searchBatch API

Listings

 Alibaba CloudRegions: 🇯🇵🇸🇬+4 ap-northeast-1, ap-southeast-1, cn-beijing, cn-hongkong, eu-central-1, us-east-1qwen3.7-maxSources: Alibaba Cloud Model Studio (DashScope) Multi-Region API, Alibaba Cloud Qwen, LiteLLM data, Models.dev data992K$2.50cache read $0.50$7.50cache write $3.125—
 Merge Gatewayqwen/qwen3.7-maxSource: Models.dev data1M$0.825cache read $0.165$2.4755cache write $1.0313—
 Ppioqwen/qwen3.7-maxSource: PPIO data1M—
 Llmgatewayqwen3.7-maxSource: LLMGateway data1M$1.25cache read $0.25$3.75—
 Zenmuxqwen/qwen3.7-maxSources: Models.dev data, ZenMux data1M$1.25cache read $0.125$3.75cache write $1.5625$0.005
 Cloudflare AI Gatewayalibaba/qwen3.7-maxSource: Models.dev data1M$1.25cache read $0.25$3.75—
 Novita AIqwen/qwen3.7-maxSources: LiteLLM data, Models.dev data, Novita AI data1M$1.25cache read $0.25$3.75—
 Orcarouterqwen/qwen3.7-maxSource: Models.dev data1M$1.25cache read $0.25$3.75cache write $1.563—
 NanoGPTqwen/qwen3.7-maxSources: Models.dev data, NanoGPT1M$1.475cache read $0.295$4.425cache write $1.8438—
 Nous Portalqwen/qwen3.7-maxSource: Nous Portal Inference Gateway data1M$1.475cache read $0.295$4.425cache write $1.8437—

4 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Agentic

Arena AgentSource: LMArena data
BFCL-V42 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · published setting
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026
JobBenchReported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MCP Atlas2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · public set · Gemini 2.5 Pro judge
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
MCP-MarkSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · published setting
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
SWE-Bench Pro3 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · internal bash + file-edit agent scaffold · temp 1.0 · top_p 0.95 · 200K · corrected/refined tasks
Qwen表:Claude Code · temp 1.0 · top_p 0.95 · 256K · refined tasks
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Terminal-Bench3 related benchmarks · 3 results
Terminal-Bench 2.0
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
2.0 · Terminus-2
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · Harbor/Terminus-2 · 5h · 12 CPU/24 GB · temp 1.0 · top_p 0.95 · top_k 20 · max_tokens 80K · 256K · avg@5
Terminal-Bench 2.1
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen表:Qwen 采用 Claude Code avg@10、5h、max_tokens 131072;友商取公开最高分

Coding

Claw-EvalSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
FrontierSWEQwen表:Claude Code;对照为 2026-08-03 官方榜 MEAN@5 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
LiveCodeBenchQwen3.7-Max release · recommended reasoning-effort=xhigh system prompt · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MLS-Bench-LiteQwen表:Claude Code · 5h · max_tokens 131072;对照来自官方榜 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
NL2RepoSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
PaperBenchQwen表:BasicAgent · Code-Dev · Claude Opus 4.6 judge · 3次均值 · 最长12h · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
QwenSVGSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
QwenSVGBench2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · QwenSVG · BT/Elo
Qwen内部双语 SVG 基准 · 自动渲染 + 多模态评判 · BT/Elo
QwenSWEBenchQwen内部基准:Claude Code · avg@3 · 8h · 256K · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
QwenWebDevQwen3.7-Max release · internal bilingual frontend benchmark · auto-render + multimodal judge · BT/Elo · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SciCode2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · published setting
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
SWE-bench3 related benchmarks · 5 results
Verified
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · internal bash + file-edit agent scaffold · temp 1.0 · top_p 0.95 · 200K
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Multilingual
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · internal bash + file-edit agent scaffold · temp 1.0 · top_p 0.95 · 200K
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
ValsSource: Vals AI

General

MAXIFE2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · English + multilingual prompts · 23 settings
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
NOVA-632 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Reasoning

ApexQwen3.7-Max release · recommended reasoning-effort=xhigh system prompt · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
AutomationBenchQwen表:600题公开子集 · Pass@1 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
CoWorkBench5 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7 release cross-harness chart
Qwen3.7-Max release · OpenClaw · release table and cross-harness figure
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Global PIQA2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · published setting
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · internal 7-domain world-model benchmark · rubric judge grounded in environment feedback
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Toolathlon VerifiedQwen表:Verified · Pass@1 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
WideSearchQwen表:友商=Claude Code,Qwen=Qwen-Agent · 4次平均 item-F1 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Math

HMMT 2026 FebQwen3.7-Max release · recommended reasoning-effort=xhigh system prompt · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
PolyMATH2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Long context

LongBench v2Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
OpenAI MRCR v2 · 8-needleQwen3.7-Max release · 128K · 8 needles · Google DeepMind eval_hub protocol · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Multilingual

EKI EstonianSource: Keelemudelite mõõdupuu data docs
MMMLU2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
WMT24++2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · 55-language average · XCOMET-XXL
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Science

CritPT2 results
Source: Epoch AI Benchmarking Hub data
Setting not stated
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond4 results
Source: Epoch AI Benchmarking Hub data
Setting not stated
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: OpenRouter benchmark exports · Sep 2026
AtlasCloud · 20 May 2026 version · 90-day results
SuperGPQA2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Human preference

Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Knowledge

Include2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
MMLU-Pro3 related benchmarks · 6 results
MMLU-Pro
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
MMLU-Redux
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
X
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · 29-language average
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Instruction following

IFBench3 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
IFEval2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Vision

VITA-Bench2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · subdomain average · Claude 4.5 Sonnet judge
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Factuality

Speech

Other

AndroidBenchQwen表:95题公开子集 · avg@3 · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
DeepSWE v1.1Qwen表:Claude Code / mini-SWE-agent 取较高值 · 256K · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
E-Commerce Bench7 related benchmarks · 7 results
Mean final assets
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
AnchorRatio · repeat-order pricing
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
BadSpend · fraudulent-supplier spend
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
Controllable return
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
CSE+ · honest-deal negotiation
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
Drawdown / peak assets
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
Show 1 more resultHide 1 result
Yuan per tool call
Reported by E-Commerce Bench authors / Alibaba Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
benchmark-owner 365-day merchant simulation · five independent episodes · up to four stores · initial balance ¥100,000 · thinking enabled where supported; effort not specified · exact API snapshot not stated for unversioned labels
HealthBenchReported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Horangi36 related benchmarks · 36 results
Kernel Bench L3 · Median SpeedupQwen3.7-Max release · 50 problems · isolated H100 80GB container · CUTLASS/CUDA docs only · 500 tool-call cap · median per-problem speedup over PyTorch eager · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Kernel Bench L3 · Win RateQwen3.7-Max release · 50 problems · isolated H100 80GB container · CUTLASS/CUDA docs only · 500 tool-call cap · fraction faster than torch.compile · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MCPMark · Version UnspecifiedQwen3.7-Max release · GitHub MCP v0.30.3 · Playwright responses truncated at 32K · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MRCR v2 · 256KQwen表:MRCR v2 · 256K · 8-needle · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
NL2Repo-Bench2 results
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7-Max release · Claude Code · repository-download commands disabled
Qwen表:Claude Code · 禁止访问目标仓库的 Bash 命令
PLawBenchQwen表:Gemini 3.1 Pro Preview judge · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
PRBench-FinanceReported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Qwen3.7 release cross-harness chart
Qwen3.7-Max release · OpenClaw · release table and cross-harness figure
QwenQoderBenchQwen内部基准:Claude Code · avg@5 · 6h · 256K · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
QwenReactBenchQwen内部双语 React 基准 · Claude Code harness · 自动渲染 + 多模态评判 · BT/Elo · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SkillsBench 1.1 · With SkillsQwen他测:v1.1 · 87 tasks · 每题3次均值;Opus/Fable=Claude Code,GPT=Codex,Qwen=OpenCode · Reported by Qwen · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · Data by LLM Stats · BullshitBench by Peter Gostev, MIT (licence) · Keelemudelite mõõdupuu: Independent LLM leaderboard for Estonian. Eesti Keele Instituut. https://moodupuu.eki.ee, CC BY 4.0. Latest per-benchmark results selected with partial-run flags retained. (licence) · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence)

Around the web

Also known as: alibaba/qwen3.7-max, bailian/qwen3.7-max, cline-pass/qwen3.7-max, dashscope/qwen3.7-max, deepinfra/qwen/qwen3.7-max, empiriolabs/qwen3-7-max, gmi/qwen/qwen3.7-max, novita/qwen/qwen3.7-max, novita/qwen3.7-max, qwen 3.7 max, qwen-3-7-max, qwen-3.7-max, qwen/qwen3.7-max, qwen/qwen3.7-max:thinking, qwen3-7-max, qwen3.7-max-2026-05-17, qwen3.7-max-2026-05-20, qwen3.7-max-2026-06-08, qwen3.7-max-20260517, qwen_ai_platform/qwen3.7-max, qwencloud/qwen3.7-max, together_ai/qwen/qwen3.7-max

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works