Qwen3.5-35B-A3B Benchmark Details

Qwen3.5-35B-A3B currently shows benchmark results led by IF Bench (50 / 282, score 72.50), τ²-Bench - Telecom (69 / 264, score 89.20), Claw Bench (8 / 29, score 91.40).

Benchmark Results

Qwen3.5-35B-A3B

Benchmark Results

General Knowledge

4 evaluations

Benchmark / mode

Score

Rank/total

HLE

Standard Mode

13.40

355 / 565

HLE

Thinking Mode

281 / 565

CritPt

Standard Mode

0.60

169 / 201

CritPt

Thinking Mode

0.90

159 / 201

Other

2 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Standard Mode

81.90

207 / 463

GPQA Diamond

Thinking Mode

84.50

168 / 463

Agent Level Benchmark

5 evaluations

Benchmark / mode

Score

Rank/total

τ²-Bench - Telecom

Standard ModeTools

86.30

82 / 264

τ²-Bench - Telecom

Thinking ModeTools

89.20

69 / 264

Terminal Bench Hard

Standard ModeTools

10.60

182 / 244

Terminal Bench Hard

Thinking ModeTools

26.50

123 / 244

τ³-Banking

Standard ModeTools

4.90

158 / 167

Instruction Following

2 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Standard Mode

44.50

183 / 282

IF Bench

Thinking Mode

72.50

50 / 282

Text Embedding

2 evaluations

Benchmark / mode

Score

Rank/total

Context Arena

Standard Mode

33.10

111 / 126

Context Arena

Thinking Mode

67.48

68 / 126

Long Context

2 evaluations

Benchmark / mode

Score

Rank/total

AA-LCR

Standard Mode

125 / 171

LongBench v2

Standard Mode

10 / 14

Claw-style Agent Evaluation

2 evaluations

Benchmark / mode

Score

Rank/total

Claw Bench

Thinking ModeTools

91.40

8 / 29

Pinch Bench

Thinking ModeTools

78.40

27 / 38

AI Agent - Tool Usage

1 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench 2.1

Standard ModeTools

40.80

147 / 194

Productivity Knowledge

1 evaluations

Benchmark / mode

Score

Rank/total

GDPval-AA v2

Standard ModeTools

740

101 / 106

Multimodal Understanding

2 evaluations

Benchmark / mode

Score

Rank/total

MMMU-Pro

Standard Mode

69.20

134 / 229

MMMU-Pro

Thinking Mode

72.70

115 / 229

Compare with other models