BenchAtlas

Rankings / Alibaba

Qwen3 5 35B A3b

open weights

released 2026-02-24

BenchAtlas Index

as of 2026-07-14
49.7
thinking
rank #220
7 families · 4 categories · medium
42.4
no reasoning
rank #277
7 families · 4 categories · medium

Benchmark evidence

40 results
agentic coding
τ²-Benchthinking89.2 %independentArtificial Analysis
τ²-Benchno reasoning86.3 %independentArtificial Analysis
coding
SciCodethinking37.7 %independentArtificial Analysis
SciCodeno reasoning29.3 %independentArtificial Analysis
external indices
Intelligence Index v4.1thinking29.3 pointsindependentArtificial Analysis
Intelligence Index v4.1no reasoning23.4 pointsindependentArtificial Analysis
human preference
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1460.6
1445.0–1476.2
communityLMArena Leaderboard Dataset
Coding (style control)older version
arena=text · category=coding · style_control=true
1434.6
1427.3–1441.9
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1432.2
1418.9–1445.5
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1430.9
1424.7–1437.2
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1423.4
1416.5–1430.3
communityLMArena Leaderboard Dataset
French (style control)older version
arena=text · category=french · style_control=true
1423.3
1401.2–1445.5
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1419.1
1407.1–1431.2
communityLMArena Leaderboard Dataset
Industry Life And Physical And Social Science (style control)older version
arena=text · category=industry_life_and_physical_and_social_science · style_control=true
1415.2
1406.3–1424.0
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1413.0
1407.7–1418.3
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1411.6
1405.9–1417.4
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1407.2
1391.9–1422.4
communityLMArena Leaderboard Dataset
Industry Business And Management And Financial Operations (style control)older version
arena=text · category=industry_business_and_management_and_financial_operations · style_control=true
1402.5
1394.3–1410.6
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1402.0
1395.4–1408.5
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1400.9
1387.9–1413.9
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1399.1
1385.0–1413.2
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1395.6
1391.3–1399.9
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1394.7
1386.0–1403.4
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1388.0
1381.2–1394.6
communityLMArena Leaderboard Dataset
Spanish (style control)older version
arena=text · category=spanish · style_control=true
1386.7
1365.7–1407.8
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1377.3
1366.3–1388.3
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1375.5
1370.0–1381.0
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1375.4
1369.4–1381.4
communityLMArena Leaderboard Dataset
Polish (style control)older version
arena=text · category=polish · style_control=true
1373.7
1349.2–1398.1
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
1365.6
1357.9–1373.3
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
1345.1
1336.7–1353.4
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1343.4
1334.1–1352.7
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond
implementation=artificial-analysis
thinking84.5 %independentArtificial Analysis
GPQA Diamond
implementation=artificial-analysis
no reasoning81.9 %independentArtificial Analysis
Humanity's Last Exam
implementation=artificial-analysis
thinking19.7 %independentArtificial Analysis
Humanity's Last Exam
implementation=artificial-analysis
no reasoning12.8 %independentArtificial Analysis
long context instruction
AA-LCRthinking62.7 %independentArtificial Analysis
AA-LCRno reasoning55.3 %independentArtificial Analysis
IFBenchthinking72.5 %independentArtificial Analysis
IFBenchno reasoning44.5 %independentArtificial Analysis

Agent + model results

systems, not bare-model scores
agent + model Artificial Analysis harness + Qwen3 5 35B A3bTerminal-Bench Hard26.5 %independentArtificial Analysis
agent + model Artificial Analysis harness + Qwen3 5 35B A3bTerminal-Bench Hard10.6 %independentArtificial Analysis

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.