Qwen3 Max (Preview)
Sep 2025
Qwen3 Max (Preview)Compare inon
Benchmarks of Qwen3 Max (Preview)32
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding1
Writing18
Health & law1
Documents1
Legacy · saturated or no longer reported4
No benchmark matches
Rankingin non-thinking mode?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperMiMo v2.6 ProunstatedOpenwins 59%-64%$0.43 / $0.87Pass back reasoning_content+3
BetterGPT-5.6 Solxhighwins 52%+426%$1.18CheaperMiMo v2.5 ProunstatedOpenwins 55%-64%$0.43 / $0.87CheaperMiMo v2.6 FlashunstatedOpenwins 55%-89%$0.14 / $0.28Pass back reasoning_content+3
CheaperGLM 5.3 FlashunstatedOpenwins 55%-84%$0.15 / $0.50Defaults to max · Self-hosted+4ModelScoreCostPer taskWhat changes
MiMo v2.6 ProunstatedOpenwins 59%-64%$0.43 / $0.87Pass back reasoning_content+3MiMo v2.5 ProunstatedOpenwins 55%-64%$0.43 / $0.87MiMo v2.6 FlashunstatedOpenwins 55%-89%$0.14 / $0.28Pass back reasoning_content+3GLM 5.3 FlashunstatedOpenwins 55%-84%$0.15 / $0.50Defaults to max · Self-hosted+4
Gemini 3 Flash Previewunstatedwins 54%-26%$0.50 / $3ModelScoreCostPer taskWhat changes
GPT-5.6 Solxhighwins 52%+426%$1.18GLM 5.2maxOpenwins 55%+41%$1.28
Muse Spark 1.2xhighwins 56%+31%$1.29Only auto tool_choice is supported+1GPT-5.5highwins 54%+639%$1.40Prompts with >272K input tokensGemini 3.6 Flashhighwins 56%same*$1.60ModelScoreCostPer taskWhat changes
MiMo v2.6 ProunstatedOpenwins 59%-64%$0.43 / $0.87Pass back reasoning_content+3GLM 5.3maxOpenwins 56%+41%$2.01Reasoning cannot be disabled+3
Kimi K3maxOpenwins 56%+294%$2.00Sampling parameters are fixed+3MiMo v2.5 ProunstatedOpenwins 55%-64%$0.43 / $0.87GLM 5.1unstatedOpenwins 55%+41%$1.40 / $4.40Context: 262K → 203KModelScoreCostPer taskWhat changes
Qwen 3.7 Max Previewwins 54%—$2.50 / $7.50Context: not verifiedQwen3.7 Plusunstatedwins 52%—$0.32 / $1.28Qwen3.6 Max Previewunstatedsame—$1.30 / $7.80Include the word JSON · QwenCloudQwen3.8 27BunstatedOpensame—$0.42 / $2.55max_tokens range · QwenCloud+4Qwen3.6 Plusunstatedsame—$0.50 / $3Qwen3.5 397B A17BunstatedOpensame—$0.60 / $3.60Qwen3 235B A22B Thinking 2507unstatedOpensame—$0.23 / $2.30No strict JSONQwen 3.5 122B A10BunstatedOpensame—$0.40 / $3.20All Alibaba (Qwen) models →History
7 versionsBenchmarks
all categoriesChat & Knowledge
Humanity's Last Exam10.1#191
AA Omniscience Index-42.8#185
CritPt (physics)0.9#132
AA Intelligence Index 4.3.2≈12.6#177
AA Omniscience · Accuracy24.2#125
GPQA Diamond76.4#140
Coding
LiveCodeBench · AA archived65.1#63
Agents & Tool Use
τ²-bench · AA32.7#143
Writing & Language
IFBench (instructions)48#124
FAQ
How much does Qwen3 Max (Preview) cost?›
$0.87 per 1M input tokens and $3.48 per 1M output tokens (cheapest host: Qiniu).
Is Qwen3 Max (Preview) free to use?›
There is no free API; the cheapest paid price is $0.87 per 1M input tokens.
How good is Qwen3 Max (Preview), and what is it best at?›
#190 of 336 in Chat & Knowledge, #93 of 239 in Coding, #141 of 242 in Agents & Tool Use, #64 of 197 in Writing & Language.
Is Qwen3 Max (Preview) better than MiniMax M3?›
On Arena Text · English, Qwen3 Max (Preview) scores 1,442.1 vs 1,442.3 — about the same.
How can I use Qwen3 Max (Preview)?›
Through the Alibaba (Qwen) API.
What is the context window of Qwen3 Max (Preview)?›
262K tokens.
When was Qwen3 Max (Preview) released? Is there a newer version?›
Released 2025-09-05. It has been superseded by Qwen3 Max.
Is Qwen3 Max (Preview) open source?›
No, Qwen3 Max (Preview) is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
Can Qwen3 Max (Preview) read images?›
No, it accepts text only.
What is Qwen3 Max (Preview)?›
Qwen3 Max (Preview) is a closed language model from Alibaba (Qwen), released Sep 2025.
What are the best alternatives to Qwen3 Max (Preview)?›
MiMo v2.6 Pro (cheaper), GPT-5.6 Sol (better).