Hy3 Pre Benchmark Details

Hy3 Pre currently shows benchmark results led by τ²-Bench - Telecom (51 / 264, score 92.70), GPQA Diamond (137 / 461, score 86.70), Terminal Bench Hard (82 / 244, score 34.10). This page also compares it with 3 competitor models, including performance and pricing views when available.

Benchmark Results

Hy3 Pre

Benchmark Results

Knowledge Exams

2 evaluations

Benchmark / mode

Score

Rank/total

HLE

Standard Mode

446 / 568

HLE

Thinking Enabled

27.80

233 / 568

Scientific Reasoning

4 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Standard Mode

73.20

300 / 461

GPQA Diamond

Thinking Enabled

86.70

137 / 461

CritPt

Standard Mode

0.30

186 / 204

CritPt

Thinking Enabled

4.60

106 / 204

Service Workflows

2 evaluations

Benchmark / mode

Score

Rank/total

τ²-Bench - Telecom

Standard ModeTools

67.50

143 / 264

τ²-Bench - Telecom

Thinking EnabledTools

92.70

51 / 264

Instruction Following

2 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Standard Mode

166 / 282

IF Bench

Thinking Enabled

63.10

110 / 282

Agentic Development

2 evaluations

Benchmark / mode

Score

Rank/total

Terminal Bench Hard

Standard ModeTools

31.80

100 / 244

Terminal Bench Hard

Thinking EnabledTools

34.10

82 / 244

Compare with other models

Competitor Comparison

Benchmark scores for Hy3 Pre compared against top models in its class

Hy3 PreGLM 5.1 Qwen3.6-Max-Preview DeepSeek-V4-Flash

Benchmark categories:

The chart shows each model’s highest score per benchmark within the current filter. Out-of-100 benchmarks use raw heights; out-of-range benchmarks are scaled within that benchmark while labels keep the original scores.

6 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.

Benchmark	Hy3 PreCurrent	GLM 5.1	Qwen3.6-Max-Preview	DeepSeek-V4-Flash
HLE Accuracy Knowledge Exams	27.80Thinking Enabled	52.30Thinking Enabled ｜ Tools	50.20Thinking Enabled ｜ Tools	51.50Thinking Level · High ｜ Tools
CritPt Score Scientific Reasoning	4.60Thinking Enabled	4.60Thinking Enabled	3.70Thinking Enabled	7.10Thinking Level · High
GPQA Diamond Accuracy Scientific Reasoning	86.70Thinking Enabled	86.20Thinking Enabled	90.40Thinking Level · High	89.40Thinking Level · High
τ²-Bench - Telecom Accuracy Service Workflows	92.70Thinking Enabled ｜ Tools	97.70Thinking Enabled ｜ Tools	95.90Thinking Enabled ｜ Tools	95.60Thinking Level · High ｜ Tools
IF Bench Accuracy Instruction Following	63.10Thinking Enabled	76.30Thinking Enabled	76.60Thinking Enabled	79.20Thinking Level · High
Terminal Bench Hard Accuracy Agentic Development	34.10Thinking Enabled ｜ Tools	43.20Thinking Enabled ｜ Tools	43.90Thinking Enabled ｜ Tools	38.60Thinking Level · High ｜ Tools

Standard API Pricing: Hy3 Pre vs. Peer Models

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier. · USD / 1M tokens

When a context threshold exists, the charted base price only applies within these limits:

Qwen3.6-Max-Preview: Base price applies to <= 128

Model	Supplier	Standard input	Standard output	Base price applies to
GLM 5.1	智谱AI	$1.4 / 1M tokens	$4.4 / 1M tokens	—
Qwen3.6-Max-Preview	阿里巴巴	$1.3 / 1M tokens	$7.8 / 1M tokens	<= 128
DeepSeek-V4-Flash	DeepSeek-AI	$0.14 / 1M tokens	$0.28 / 1M tokens	—