Qwen3.5-35B-A3B

Qwen3.5-35B-A3B is a reasoning model from Alibaba. 21 benchmarks count toward its score, in 7 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.49.7 ±4.0
CoverageShare of the index weight with results.95%
SpeedOutput tokens per second.36/s
Input / 1MUS dollars per 1M input tokens.$0.313
Output / 1MUS dollars per 1M output tokens.$1.25
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.1395 (#148)

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

29,043 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.
49.2
CodingCode writing and repair.
53.0
ReasoningLogic problems and puzzles.
44.9
MultimodalTasks with images and text.
53.0
KnowledgeFacts and expert knowledge.
49.6
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
56.0
MathMath problems.
50.7

Results

21 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.
V*Multimodal92.7%52.4Z.AI
Instruction-Following EvalInstruction91.9%49.3Jeffrey Zhou et al.
τ²-Bench Tool-Agent-User EvaluationAgentic89.2%62.8Victor Barres et al.
Massive Multitask Language Understanding ProfessionalKnowledge85.3%54.9Yubo Wang et al.
Artificial Analysis GPQA DiamondKnowledge84.5%55.2Artificial Analysis
Graduate-Level Google-Proof Q&AKnowledge84.2%56.0David Rein et al.
MathVisionMultimodal83.9%Qwen
Massive Multi-discipline Multimodal UnderstandingMultimodal81.4%50.6MMMU authors
GPQA diamondKnowledge81.2%53.2none effortEpoch AI
MMLU-ProXMultilingual81.0%MMLU-ProX authors
Artificial Analysis MMMU-ProMultimodal72.7%55.9Artificial Analysis
Artificial Analysis IFBenchInstruction72.5%62.7Artificial Analysis
Multimodal Multi-disciplinary Video UnderstandingMultimodal72.3%MMVU benchmark maintainers
Artificial Analysis Long Context ReasoningReasoning72.0%58.1Artificial Analysis
OTIS Mock AIME 2024-2025Math70.0%50.7none effortEpoch AI
Software Engineering Benchmark VerifiedCoding69.2%53.0Carlos E. Jimenez et al.
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesKnowledge63.4%49.9Xiaoxuan Du et al.
BrowseCompAgentic61.0%51.1OpenAI
LongBench v2Reasoning59.0%LongBench v2 authors
OSWorld-VerifiedAgentic54.5%44.4Tianbao Xie et al.
SWE-RebenchCoding53.7%Nebius
Gert Labs Composite Game BenchmarkAgentic29.0%38.3Gert Labs
Artificial Analysis Humanity's Last ExamKnowledge21.0%47.3Artificial Analysis
Artificial Analysis Omniscience AccuracyKnowledge20.1%38.7Artificial Analysis
Artificial Analysis Intelligence IndexKnowledge19.3%46.6Artificial Analysis
Chess PuzzlesReasoning10.0%36.5none effortEpoch AI
Critical Physics TasksReasoning0.9%40.3Artificial Analysis

21 benchmarks count, from 22 of 27 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 attribution

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