Qwen 3.7 Max Preview
May 2026
Qwen 3.7 Max PreviewCompare inon
Benchmarks of Qwen 3.7 Max Preview9
BenchmarkByScoreRank
No benchmark matches
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGemini 3.8 Flashhighwins 57%-60%*$1.24Sampling parameters are ignored+1
BetterGPT-6.1 Solmaxwins 53%+7%$0.72Remove unsupported parameters+5
OpenKimi K3maxwins 53%+60%$2.00Sampling parameters are fixed+3CheaperGemini 3.6 Flashhighwins 55%-60%*$1.60
CheaperMuse Sparkunstatedwins 54%-47%$1.25 / $4.25Only n=1 · Meta / Chat Completions+1ModelScoreCostPer taskWhat changes
Gemini 3.8 Flashhighwins 57%-60%*$1.24Sampling parameters are ignored+1Gemini 3.6 Flashhighwins 55%-60%*$1.60Muse Sparkunstatedwins 54%-47%$1.25 / $4.25Only n=1 · Meta / Chat Completions+1Muse Spark 1.3maxwins 53%-47%$1.60Only auto tool_choice is supported+5Gemini 3 Flash Previewunstatedwins 53%-70%$0.50 / $3ModelScoreCostPer taskWhat changes
GPT-6.1 Solmaxwins 53%+7%$0.72Remove unsupported parameters+5GPT-5.6 Solxhighwins 54%+113%$1.18Gemini 3.8 Flashhighwins 57%-60%*$1.24Sampling parameters are ignored+1
GLM 5.2maxOpenwins 52%-43%$1.47Gemini 3.6 Flashhighwins 55%-60%*$1.60ModelScoreCostPer taskWhat changes
Kimi K3maxOpenwins 53%+60%$2.00Sampling parameters are fixed+3GLM 5.2maxOpenwins 52%-43%$1.47GLM 5.3maxOpensame-43%$2.01Reasoning cannot be disabled+3GLM 5unstatedOpensame-59%$1 / $3.20GLM 5.1unstatedOpensame-43%$1.40 / $4.40ModelScoreCostPer taskWhat changes
Qwen3.6 Max Previewunstatedsame—$1.30 / $7.80Include the word JSON · QwenCloudQwen3.7 Plusunstatedsame—$0.32 / $1.28Qwen3.6 Plusunstatedwins 45%—$0.50 / $3Qwen3.5 397B A17BunstatedOpenwins 45%—$0.60 / $3.60Qwen3 Max (Preview)unstatedwins 43%—$0.87 / $3.48Qwen 3Openwins 43%——Qwen3 235B A22B Thinking 2507unstatedOpenwins 40%—$0.23 / $2.30No strict JSONQwen3.8 27BunstatedOpenwins 40%—$0.42 / $2.55max_tokens range · QwenCloud+4All Alibaba (Qwen) models →History
2 versionsBenchmarks
all categoriesChat & Knowledge
Arena Text · overall1,474#27
Arena Text · Chinese1,529.6#13
Arena Text · English1,469.9#28
Arena Text · non-English1,473.9#13
Arena Text · Russian1,483.3#14
Arena Text · overall · full cohort1,475.2#16
Writing & Language
Arena Text · creative writing1,439#37
FAQ
How much does Qwen 3.7 Max Preview cost?›
$2.50 per 1M input tokens and $7.50 per 1M output tokens.
Is Qwen 3.7 Max Preview free to use?›
There is no free API; the cheapest paid price is $2.50 per 1M input tokens.
How good is Qwen 3.7 Max Preview, and what is it best at?›
#27 of 196 in Chat & Knowledge, #37 of 195 in Writing & Language.
Is Qwen 3.7 Max Preview better than Grok 4.6?›
On Arena Text · creative writing, Qwen 3.7 Max Preview scores 1,439 vs 1,442 — about the same.
How can I use Qwen 3.7 Max Preview?›
Through the Alibaba (Qwen) API.
When was Qwen 3.7 Max Preview released? Is there a newer version?›
Released 2026-05-20. It is the current version in its line.
Can Qwen 3.7 Max Preview read images?›
No, it accepts text only.
What is Qwen 3.7 Max Preview?›
Qwen 3.7 Max Preview is a language model from Alibaba (Qwen), released May 2026.
What are the best alternatives to Qwen 3.7 Max Preview?›
Gemini 3.8 Flash (cheaper), GPT-6.1 Sol (better), Kimi K3 (open weights).