Qwen3.5-27B

Qwen3.5-27B is a reasoning model from Alibaba. 19 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.53.2 ±5.2
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.34/s
Input / 1MUS dollars per 1M input tokens.$0.195
Output / 1MUS dollars per 1M output tokens.$1.56
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.1408 (#131)

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

27,227 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.
52.7
CodingCode writing and repair.
55.6
ReasoningLogic problems and puzzles.
51.1
MultimodalTasks with images and text.
54.6
KnowledgeFacts and expert knowledge.
51.8
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
60.9
MathMath problems.
N/A

Results

19 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.
Instruction-Following EvalInstruction95.0%56.0Jeffrey Zhou et al.
τ²-Bench Tool-Agent-User EvaluationAgentic93.9%66.2Victor Barres et al.
V*Multimodal93.7%53.7Z.AI
Massive Multitask Language Understanding ProfessionalKnowledge86.1%56.2Yubo Wang et al.
MathVisionMultimodal86.0%Qwen
Artificial Analysis GPQA DiamondKnowledge85.8%56.5Artificial Analysis
Graduate-Level Google-Proof Q&AKnowledge85.5%57.2David Rein et al.
Massive Multi-discipline Multimodal UnderstandingMultimodal82.3%51.5MMMU authors
MMLU-ProXMultilingual82.2%MMLU-ProX authors
Artificial Analysis Long Context ReasoningReasoning77.7%62.0Artificial Analysis
Artificial Analysis IFBenchInstruction75.6%65.8Artificial Analysis
Artificial Analysis MMMU-ProMultimodal75.0%58.7Artificial Analysis
Multimodal Multi-disciplinary Video UnderstandingMultimodal73.3%MMVU benchmark maintainers
Software Engineering Benchmark VerifiedCoding72.4%55.6Carlos E. Jimenez et al.
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesKnowledge65.6%51.8Xiaoxuan Du et al.
BrowseCompAgentic61.0%51.1OpenAI
LongBench v2Reasoning60.6%LongBench v2 authors
SWE-RebenchCoding58.9%Nebius
OSWorld-VerifiedAgentic56.2%46.0Tianbao Xie et al.
Gert Labs Composite Game BenchmarkAgentic39.4%47.5Gert Labs
Artificial Analysis Humanity's Last ExamKnowledge23.9%50.5Artificial Analysis
Artificial Analysis Intelligence IndexKnowledge22.9%51.0Artificial Analysis
Artificial Analysis Omniscience AccuracyKnowledge20.7%39.5Artificial Analysis
Critical Physics TasksReasoning0.9%40.3Artificial Analysis

19 benchmarks count, from 19 of 24 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 stated

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