Qwen 3.6 Max (preview)

Qwen 3.6 Max (preview) is a reasoning model from Alibaba in the Qwen 3.6 Max family. 22 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.58.3 ±4.9
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.64/s
Input / 1MUS dollars per 1M input tokens.$1.03
Output / 1MUS dollars per 1M output tokens.$6.16
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.1446 (#50)

The index is a score out of 100. The ± range shows how much it can change.

5,186 votes. Elo shows what people prefer. It does not change the score.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
61.7
CodingCode writing and repair.
58.2
ReasoningLogic problems and puzzles.
53.4
MultimodalTasks with images and text.
N/A
KnowledgeFacts and expert knowledge.
60.5
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
66.8
MathMath problems.
54.2

Results

22 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
τ²-Bench Tool-Agent-User EvaluationAgentic95.9%67.6Victor Barres et al.
OTIS Mock AIME 2024-2025Math91.1%62.5Epoch AI
Artificial Analysis GPQA DiamondKnowledge88.8%59.6Artificial Analysis
GPQA diamondKnowledge87.4%58.9Epoch AI
Artificial Analysis Long Context ReasoningReasoning80.7%64.1Artificial Analysis
SWE-Bench verifiedCoding76.7%59.0Epoch AI
Artificial Analysis IFBenchInstruction76.6%66.8Artificial Analysis
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesKnowledge73.9%58.7Xiaoxuan Du et al.
SWE-benchCoding72.8%55.91 Sept 2026Vals AI
QwenClawBenchAgentic59.0%57.9Qwen
SWE-bench ProCoding57.3%59.4Xiang Deng et al.
SimpleQA VerifiedKnowledge52.0%69.7Epoch AI
Terminal-Bench 2.0Agentic51.7%59.64 Jun 2026Vals AI
Scientific Code BenchmarkCoding47.0%57.4Benchmark authors
NL2RepoCoding42.9%58.6MiniMax
Artificial Analysis Omniscience AccuracyKnowledge37.9%60.8Artificial Analysis
Artificial Analysis Humanity's Last ExamKnowledge30.8%58.0Artificial Analysis
Artificial Analysis Intelligence IndexKnowledge28.4%57.9Artificial Analysis
FrontierMath-2025-02-28-PrivateMath23.1%52.2Epoch AI
Chess PuzzlesReasoning20.0%49.4Epoch AI
Mystery Game PuzzlesReasoning19.0%54.1none effortEpoch AI
FrontierMath-Tier-4-2025-07-01-PrivateMath4.2%48.0Epoch AI
Critical Physics TasksReasoning3.7%46.1Artificial Analysis

22 benchmarks count, from 23 of 23 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authorsOpenRouter, collected directlyNo licence statedEpoch AI, collected directlyCC BY — free to use and redistribute with attributionVals AI, collected directlyNo licence stated. Read from the public leaderboard and credited to Vals AI

Same level, lower price

Qwen3.8-Flash-Next66.4 · Freedots3-note Preview65.5 · FreeApodex 1.1 Mini59.1 · Free

More from Alibaba

Qwen3.8 Max68.4Qwen3.8-Flash-Next66.4Qwen3.8 Max Preview69.0Qwen3.8-27B63.1Qwen3.7 Max62.6