Inference
EN
Alibaba

Qwen3.7 Plus

Qwen3.7 Plus is a reasoning model from Alibaba. 48 benchmarks count toward its index, in 8 categories.

  • Closed weights
  • Reasoning
  • Input: text · image
  • Released 2026-06
RankedEnough results for an index.
IndexAll categories, out of 100.57.7±2.3Median 51.0
CoverageIndex weight that has results.100%
SpeedOutput tokens per second.51/sMedian 67/s
First tokenSeconds to the first output token.3.5 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.32batch $0.23Median $0.60
Output / 1MUS dollars per 1M output tokens.$1.28batch $0.92 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.1454±4 #47Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
qwen/qwen3.7-plus
Max outputMaximum output tokens per request.
131K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.064
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
1 provider
ProviderTypeMaker, cloud, host or router.InputOutputContext
OpenRouterRouter$0.32$1.281M

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.15 counted57.387 / 218
CodingCode writing and repair.8 counted58.378 / 209
ReasoningLogic problems and puzzles.4 counted57.774 / 227
MultimodalTasks with images and text.7 counted62.635 / 84
KnowledgeFacts and expert knowledge.6 counted54.897 / 268
MultilingualTasks in many languages.2 counted54.04 / 9
InstructionTasks with strict prompt rules.2 counted58.118 / 63
MathMath problems.4 counted55.462 / 168

Results

48 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Massive Multitask Language Understanding ProfessionalKnowledge88.5%3 / 75
Artificial Analysis IFBenchInstruction78.0%8 / 146
SimpleVQAMultimodal81.7%1 / 18
Terminal-Bench 2.0 · BenchLMAgentic70.3%4 / 41
Instruction-Following EvalInstruction94.6%7 / 70
VITA-BenchAgentic45.6%2 / 13
Artificial Analysis MMMU-ProMultimodal80.5%18 / 113
LiveCodeBench v6Coding89.6%10 / 59
MMLU-ProXMultilingual85.4%3 / 17
Artificial Analysis GPQA DiamondKnowledge90.0%36 / 199
Graduate-Level Google-Proof Q&AKnowledge90.3%28 / 141
MMMLUKnowledge89.0%3 / 15
QwenClawBenchAgentic61.8%2 / 10
Scientific Code BenchmarkCoding51.3%7 / 35
MMLU-ReduxKnowledge94.5%4 / 19
MRCRv2Reasoning91.7%3 / 14
τ²-Bench Tool-Agent-User EvaluationAgentic93.0%33 / 149
Harvard-MIT Mathematics Tournament February 2026Math92.9%6 / 27
RealWorldQAMultimodal86.9%4 / 18
Berkeley Function Calling Leaderboard v4Agentic72.9%6 / 26
Software Engineering Benchmark VerifiedCoding77.7%31 / 114
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesKnowledge71.4%6 / 22
Artificial Analysis Humanity's Last ExamKnowledge35.6%61 / 222
MathVisionMultimodal90.3%7 / 24
Critical Physics TasksReasoning9.1%70 / 214
GPQA Diamond · BenchLMKnowledge90.3%28 / 84
Instruction Following BenchmarkInstruction79.1%19 / 57
Massive Multi-discipline Multimodal Understanding ProMultimodal79.0%19 / 53
CharXiv ReasoningMultimodal85.9%18 / 49
GPQA diamond · Epoch AIKnowledge81.8% ±2.782 / 217
ERQAMultimodal69.8%7 / 18
SkillsBenchCoding54.3% ±4.314 / 35
OTIS Mock AIME 2024-2025Math80.0% ±6.086 / 200
ScreenSpot ProMultimodal79.0%9 / 20
Artificial Analysis Long Context ReasoningReasoning73.0%103 / 217
IMOAnswerBenchMath86.0%5 / 10
AndroidWorldAgentic81.0%5 / 10
SWE-bench ProCoding57.6%42 / 83
OSWorld-VerifiedAgentic73.3%19 / 37
Terminal-Bench HardAgentic47.0%12 / 23
MCP AtlasAgentic73.2%28 / 52
Humanity's Last ExamKnowledge34.7%42 / 77
Claw-EvalAgentic62.7%23 / 42
INCLUDEMultilingual83.0%7 / 12
Vibe Code Bench v1.1Coding46.4% ±4.674 / 123
Artificial Analysis SciCodeCoding46.1%69 / 114
APEX-Agents-AAAgentic22.4%19 / 30
Mystery Game PuzzlesReasoning17.0% ±3.851 / 80
GDPval-AA normalizedAgentic13.3%90 / 133
FrontierMath-Tiers-1-3-v2-PrivateMath34.4% ±2.863 / 90
NL2RepoCoding41.1%26 / 36
Terminal-Bench 2.1 · Vals AIAgentic52.8% ±0.657 / 75
Code MigrationCoding12.9% ±2.961 / 77
Agent Arena steerabilityAgentic−11.645 / 54
Agent Arena task outcomeAgentic−10.846 / 54
Agent Arena command recoveryAgentic−6.343 / 50
OSWorld 2.0Agentic2.8%25 / 25
SAGEEducation89.1% ±1.46 / 29
MortgageTaxFinance66.2% ±0.936 / 97
EMBFinance49.3% ±3.256 / 75
Legal Research BenchLegal16.3% ±2.669 / 90
Finance Agent (v2)Finance38.2% ±1.064 / 79
Tax Agent BenchFinance10.5% ±2.067 / 82
OmniDocBench 1.5Multimodal91.4%—
MAXIFEMultilingual88.8%—
Video-MME with subtitleMultimodal88.0%—
MLVU mean averageMultimodal87.4%—
VideoMMMUMultimodal85.4%—
PolyMathMultilingual84.0%—
MedXpertQA MultimodalMultimodal71.0%—
OCRBench V2Multimodal70.7%—
DeepPlanningAgentic62.3%—
NOVA-63Multilingual58.8%—
ODINW13Multimodal51.1%—
MMSearch-PlusMultimodal41.4%—
Artificial Analysis EnterpriseOps-GymAgentic40.6%—
ApexMath22.7%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYVals AINo licence statedLMArenaCC BY 4.0models.devMIT

History

5 releases
Index58.5 → 57.7
23 Oct 202612 Nov 2026
Output / 1M$1.28 → $1.28
23 Oct 202612 Nov 2026

Questions

How much does Qwen3.7 Plus cost?
$0.32 per 1M input tokens, $1.28 per 1M output tokens.
How good is Qwen3.7 Plus?
Its index is 57.7 out of 100. That is place 86 of 283 ranked models.
What is Qwen3.7 Plus best at?
Instruction: 58.1 out of 100, place 18 of 63.
How fast is Qwen3.7 Plus?
51 output tokens per second, 3.5 s to the first token.
What is the context window of Qwen3.7 Plus?
1M tokens.
Is Qwen3.7 Plus open source?
No. Its weights are not published.
When was Qwen3.7 Plus released?
June 2026.

Badge

Qwen3.7 Plus on Inference360: place 86 of 283, index 57.7
HTML
<a href="https://inference360.ai/models/qwen3-7-plus"><img src="https://inference360.ai/badges/qwen3-7-plus.svg" alt="Qwen3.7 Plus on Inference360: place 86 of 283, index 57.7" height="20"></a>
Markdown
[![Qwen3.7 Plus on Inference360: place 86 of 283, index 57.7](https://inference360.ai/badges/qwen3-7-plus.svg)](https://inference360.ai/models/qwen3-7-plus)

Compare with

Kimi K2.7 Code57.7DeepSeek V4 Pro 042357.6Gemini 3 Flash Preview57.5Gemini 3 Pro57.9Qwen3.7 Max60.5

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeGLM-5.3-Flash63.1 · $0.078/1M

More from Alibaba

All models
Qwen3.7 Max60.5Qwen-AgentWorld-35B-A3BUnrankedQwen3-ASR 0.6BUnrankedQwen3-ASR 1.7BUnrankedQwen3.6 FlashUnranked