LLM Reference

Qwen3-Max

Released
2025-04-28
Last refreshed
2026-06-29
Status
Researched 137d ago
Open sourceCommercial use: permittedMultimodalCodingRAGAgentsLong contextVisionJSON / Tool use

Qwen3-Max is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 262k context window
  • Buyers comparing 3 tracked provider routes

Do not use it for

  • Workloads where another current model has stronger sourced task evidence
Specifications
Family
Qwen3
Released
2025-04-28
Context
262k
Architecture
Decoder Only
Knowledge cutoff
2025-12
Specialization
general
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Unknown
Code
Unknown
Training
Fine-tuned
Created by
AlibabaAlibaba

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China
Founded 2017
Website
Pricing
Output / 1M
$3.90
Input / 1M
$0.780

Cheapest of 3 routes · OpenRouter

Providers(3)
OpenRouterVercel AI GatewayNovita AI
View 3 provider routes
Links
Website

About

Alibaba's Qwen3-Max, flagship model with improved multilingual and reasoning capabilities.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ D

1 relevant benchmark in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Q/$ C

4 relevant benchmarks in the decision map.

Provider price ladder

Compare all 3

Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MCacheRoute
OpenRouter$0.780$3.90-
Serverless
Vercel AI Gateway$1.20$6.00read $0.240
Serverless
Novita AI$2.11$8.45-
Serverless

Capabilities

VisionMultimodalJSON / Tool useStructured Outputs

Benchmark peer barsfor Coding

SWE-bench VerifiedRank 22 of 81
Claude Fable 5
96.0
Claude Opus 5
96.0
Claude Mythos Preview
93.9
Claude Opus 4.8
88.6
Qwen3-Maxcurrent
78.8

Benchmark scores(4)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
SWE-bench Verified78.8SWE-bench VerifiedObserved 2026-04-24—Source
τ-bench76.8τ-benchObserved 2026-04-24—Source
Berkeley Function Calling Leaderboard v371.9Berkeley Function Calling Leaderboard (BFCL v3)Observed 2026-04-12—Source
MultiChallenge41.2MultiChallengeObserved 2026-04-26—Source

Migration checks

No linked migration route is available for this model yet.

Compare Qwen3-Max with other models

Show all 31 popular comparisonssorted by 7-day search impressions
Qwen3-Max vs Grok Build 0.1140Qwen3-Max vs GPT-5.5131Qwen3-Max vs Llama 3 8B Instruct122Qwen3-Max vs Claude Sonnet 4.5116Qwen3-Max vs DeepSeek R1 Lite93Qwen3-Max vs Qwen3.5-397B-A17B90Qwen3-Max vs GLM-5.143Qwen3-Max vs Llama 3 70B Instruct41Qwen3-Max vs GPT-5 Pro39Qwen3-Max vs o326Qwen3-Max vs Llama 3.2 1B Instruct25Qwen3-Max vs Mixtral 8x7B24Qwen3-Max vs Mistral Large 223Qwen3-Max vs Gemini 2.5 Flash17Qwen3-Max vs GPT-5.216Qwen3-Max vs Grok 4.314Qwen3-Max vs Gemini 2.5 Pro11Qwen3-Max vs GPT-4o (11-20)11Qwen3-Max vs Claude Opus 4.510Qwen3-Max vs Gemini 3.1 Pro Preview Custom Tools10Qwen3-Max vs Llama 3.2 11B Instruct8Qwen3-Max vs Gemini 3.1 Flash-Lite7Qwen3-Max vs GPT-4o Audio7Qwen3-Max vs Claude Mythos Preview6Qwen3-Max vs GPT-4o-mini Search Preview6Qwen3-Max vs Grok-35Qwen3-Max vs DeepSeek R1 05284Qwen3-Max vs GPT-4o Mini (07-18)4Qwen3-Max vs o3 Mini1Qwen3-Max vs GPT-5.41Qwen3-Max vs Kimi K2.50