Qwen3.8 Max

AlibabaMCP Tool CallReasoning

alibaba/qwen3.8-maxClosed API

2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows

Benchmark Pricing$2.00 / 1M in$6.00 / 1M out$0.25 cache read

Context: 1,000,000 tokens•Max Output: 131,072 tokens•Released: 2026-08-03

TECHNICAL SPECIFICATIONS

Context Window1,000,0001M tokens

Max Output131,072131.1K tokens

LicenseProprietaryCloud API only

Input Modalities

TextVisionPDFVideo

Function / Tool CallingSupported (MCP Ready)

Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Qwen3.8 Max natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 1,000,000 tokens, this model can handle extremely large MCP resource manifests, complete database schemas, and multiple active server tool definitions simultaneously without degradation.

STANDARDIZED BENCHMARKS & EVALS

Terminal-Benchavg@10

86.6

SWE-Bench Proresolved

67.7

DeepSWEscore

56.6

NL2Reposcore

55.9

FrontierSWEdominance score

73.5

MLS-Bench-Litescore

PaperBenchscore

AndroidBenchavg@3

75.1

QwenSWEBenchavg@3

80.7

QwenQoderBenchavg@5

58.4

QwenReactBenchElo

1724

QwenSVGBenchElo

1713

CoWorkBenchscore

74.8

WorkSpaceBenchscore

67.7

JobBenchscore

53.4

SkillsBenchavg@3

70.2

Agents' Last Exampass@1

Agents' Last Examscore

52.4

AutomationBenchpass@1

27.3

Toolathlon-Verifiedpass@1

72.5

WideSearchaverage item F1

81.9

Humanity's Last Examscore

56.2

Humanity's Last Examscore

43.6

GPQA Diamondscore

92.6

IFBenchscore

82.8

$OneMillion-Benchexpert score

52.5

HealthBenchscore

60.2

PLawBenchscore

73.2

PRBenchscore

57.6

PRBenchscore

58.3

MRCRscore

92.9

LongBenchscore

66.3

API CALL INVOCATION EXAMPLE

// Example: Calling Qwen3.8 Max via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.ALIBABA_API_KEY,
});

const response = await client.chat.completions.create({
  model: "alibaba/qwen3.8-max",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);