Model profile · Tencent

hunyuan-hy3-preview

Closed weightsmid · registry tag 2026 benchmark-derived

Thin verified coverage

Reads as thin verified coverage across the resolved source data.

Visible coverage: 2.1%
Verified coverage: 2.1%
Spread: n/a
Last verified: Jun 20, 2026

textcode1 aliases28 official source links

Open compare

Data version

Current snapshot.

Data version Jun 20, 2026Model list checked9 providers · 1081 tracked modelsPage refreshed Jul 5, 2026

The registry snapshot and page stamp are shown so a stale deploy is visible at a glance.

Source-linked scores by benchmark

Each row keeps the benchmark source, source type, raw metric, and percentile inside its fair comparison set.

Thin verified coverageThis model currently reads as thin verified coverage across the resolved source data.

Chat / text18 benchmarks73.4%

Text Arena

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #84

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,413
Percentile: 74.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: overall. Source rank: #103. Votes: 6678. Organization: tencent. License: tencent-hunyuan-community.

74.5% percentile inside its fair comparison set

1,413Raw benchmark valueCI 1,406 - 1,421

Text Arena · Creative Writing

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #105

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 67.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: creative_writing. Source rank: #130. Votes: 1019. Organization: tencent. License: tencent-hunyuan-community.

67.8% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,339 - 1,378

Text Arena · English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #89

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,420
Percentile: 72.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: english. Source rank: #109. Votes: 3234. Organization: tencent. License: tencent-hunyuan-community.

72.9% percentile inside its fair comparison set

1,420Raw benchmark valueCI 1,410 - 1,431

Text Arena · Exclude Ties

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #84

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 74.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: exclude_ties. Source rank: #103. Votes: 5028. Organization: tencent. License: tencent-hunyuan-community.

74.5% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,391 - 1,411

Text Arena · Hard Prompts

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #74

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,438
Percentile: 77.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: hard_prompts. Source rank: #92. Votes: 4422. Organization: tencent. License: tencent-hunyuan-community.

77.5% percentile inside its fair comparison set

1,438Raw benchmark valueCI 1,429 - 1,448

Text Arena · Hard Prompts English

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #85

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,438
Percentile: 74.1%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: hard_prompts_english. Source rank: #104. Votes: 2227. Organization: tencent. License: tencent-hunyuan-community.

74.1% percentile inside its fair comparison set

1,438Raw benchmark valueCI 1,425 - 1,450

Text Arena · Instruction Following

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #88

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,399
Percentile: 73.2%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: instruction_following. Source rank: #109. Votes: 2218. Organization: tencent. License: tencent-hunyuan-community.

73.2% percentile inside its fair comparison set

1,399Raw benchmark valueCI 1,386 - 1,412

Text Arena · Longer Query

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #74

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 76%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: longer_query. Source rank: #92. Votes: 2906. Organization: tencent. License: tencent-hunyuan-community.

76% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,415 - 1,438

Text Arena · Multi Turn

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #81

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 75.2%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: multi_turn. Source rank: #100. Votes: 1177. Organization: tencent. License: tencent-hunyuan-community.

75.2% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,397 - 1,433

Text Arena · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #90

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,405
Percentile: 72.6%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: overall. Source rank: #108. Votes: 6678. Organization: tencent. License: tencent-hunyuan-community.

72.6% percentile inside its fair comparison set

1,405Raw benchmark valueCI 1,397 - 1,412

Text Arena · Creative Writing · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #107

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,345
Percentile: 67.2%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: creative_writing. Source rank: #131. Votes: 1019. Organization: tencent. License: tencent-hunyuan-community.

67.2% percentile inside its fair comparison set

1,345Raw benchmark valueCI 1,326 - 1,364

Text Arena · English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #95

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,411
Percentile: 71.1%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: english. Source rank: #113. Votes: 3234. Organization: tencent. License: tencent-hunyuan-community.

71.1% percentile inside its fair comparison set

1,411Raw benchmark valueCI 1,400 - 1,421

Text Arena · Exclude Ties · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #89

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,388
Percentile: 72.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: exclude_ties. Source rank: #107. Votes: 5028. Organization: tencent. License: tencent-hunyuan-community.

72.9% percentile inside its fair comparison set

1,388Raw benchmark valueCI 1,378 - 1,398

Text Arena · Hard Prompts · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #79

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,418
Percentile: 76%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: hard_prompts. Source rank: #96. Votes: 4422. Organization: tencent. License: tencent-hunyuan-community.

76% percentile inside its fair comparison set

1,418Raw benchmark valueCI 1,409 - 1,427

Text Arena · Hard Prompts English · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #86

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,419
Percentile: 73.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: hard_prompts_english. Source rank: #103. Votes: 2227. Organization: tencent. License: tencent-hunyuan-community.

73.8% percentile inside its fair comparison set

1,419Raw benchmark valueCI 1,406 - 1,431

Text Arena · Instruction Following · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #89

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,386
Percentile: 72.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: instruction_following. Source rank: #107. Votes: 2218. Organization: tencent. License: tencent-hunyuan-community.

72.9% percentile inside its fair comparison set

1,386Raw benchmark valueCI 1,373 - 1,399

Text Arena · Longer Query · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #79

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,407
Percentile: 74.3%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: longer_query. Source rank: #97. Votes: 2906. Organization: tencent. License: tencent-hunyuan-community.

74.3% percentile inside its fair comparison set

1,407Raw benchmark valueCI 1,396 - 1,419

Text Arena · Multi Turn · No Style Control

AR · Chat / text · Human

It tests whether the model is actually useful in normal conversational turns, not just on narrow correctness tasks.

Rank #80

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,410
Percentile: 75.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: multi_turn. Source rank: #97. Votes: 1177. Organization: tencent. License: tencent-hunyuan-community.

75.5% percentile inside its fair comparison set

1,410Raw benchmark valueCI 1,392 - 1,428

Coding6 benchmarks47.5%

Code Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 35.6%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: overall. Source rank: #57. Votes: 1352. Organization: tencent. License: tencent-hunyuan-community.

35.6% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,346 - 1,380

WebDev Arena

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #48

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,363
Percentile: 35.6%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: webdev. Source rank: #57. Votes: 1352. Organization: tencent. License: tencent-hunyuan-community.

35.6% percentile inside its fair comparison set

1,363Raw benchmark valueCI 1,346 - 1,380

Code Arena · Webdev Html

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #51

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,354
Percentile: 31.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: webdev-html. Source rank: #60. Votes: 190. Organization: tencent. License: tencent-hunyuan-community.

31.5% percentile inside its fair comparison set

1,354Raw benchmark valueCI 1,310 - 1,398

Code Arena · Webdev React

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #42

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,358
Percentile: 30.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: webdev-react. Source rank: #53. Votes: 1160. Organization: tencent. License: tencent-hunyuan-community.

30.5% percentile inside its fair comparison set

1,358Raw benchmark valueCI 1,340 - 1,377

Text Arena · Coding

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #77

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,459
Percentile: 76.3%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: coding. Source rank: #97. Votes: 1974. Organization: tencent. License: tencent-hunyuan-community.

76.3% percentile inside its fair comparison set

1,459Raw benchmark valueCI 1,446 - 1,473

Text Arena · Coding · No Style Control

AR · Coding · Human

It tells you whether the model can generate, repair, and reason over code under evaluator pressure rather than marketing examples.

Rank #80

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,427
Percentile: 75.3%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: coding. Source rank: #97. Votes: 1974. Organization: tencent. License: tencent-hunyuan-community.

75.3% percentile inside its fair comparison set

1,427Raw benchmark valueCI 1,414 - 1,441

Reasoning / math / science2 benchmarks80.7%

Text Arena · Math

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #56

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 82.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: math. Source rank: #71. Votes: 405. Organization: tencent. License: tencent-hunyuan-community.

82.5% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,401 - 1,456

Text Arena · Math · No Style Control

AR · Reasoning / math / science · Human

It is one of the cleaner reads on deliberate reasoning strength rather than style or popularity.

Rank #67

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 79%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: math. Source rank: #80. Votes: 405. Organization: tencent. License: tencent-hunyuan-community.

79% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,394 - 1,449

Professional reasoning18 benchmarks75.6%

Text Arena · Expert

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #63

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,447
Percentile: 77.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: expert. Source rank: #81. Votes: 652. Organization: tencent. License: tencent-hunyuan-community.

77.5% percentile inside its fair comparison set

1,447Raw benchmark valueCI 1,424 - 1,469

Text Arena · Industry Business And Management And Financial Operations

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #74

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 77%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_business_and_management_and_financial_operations. Source rank: #91. Votes: 1374. Organization: tencent. License: tencent-hunyuan-community.

77% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,398 - 1,431

Text Arena · Industry Entertainment And Sports And Media

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #95

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,368
Percentile: 70.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #117. Votes: 1318. Organization: tencent. License: tencent-hunyuan-community.

70.9% percentile inside its fair comparison set

1,368Raw benchmark valueCI 1,351 - 1,385

Text Arena · Industry Legal And Government

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #53

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,444
Percentile: 82.6%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_legal_and_government. Source rank: #70. Votes: 527. Organization: tencent. License: tencent-hunyuan-community.

82.6% percentile inside its fair comparison set

1,444Raw benchmark valueCI 1,417 - 1,471

Text Arena · Industry Life And Physical And Social Science

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #72

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,437
Percentile: 78%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #90. Votes: 1223. Organization: tencent. License: tencent-hunyuan-community.

78% percentile inside its fair comparison set

1,437Raw benchmark valueCI 1,419 - 1,454

Text Arena · Industry Mathematical

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #43

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 86.4%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_mathematical. Source rank: #54. Votes: 423. Organization: tencent. License: tencent-hunyuan-community.

86.4% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,421 - 1,477

Text Arena · Industry Medicine And Healthcare

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #93

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,428
Percentile: 68.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_medicine_and_healthcare. Source rank: #113. Votes: 468. Organization: tencent. License: tencent-hunyuan-community.

68.8% percentile inside its fair comparison set

1,428Raw benchmark valueCI 1,400 - 1,457

Text Arena · Industry Software And It Services

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #81

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,449
Percentile: 75.4%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_software_and_it_services. Source rank: #100. Votes: 2695. Organization: tencent. License: tencent-hunyuan-community.

75.4% percentile inside its fair comparison set

1,449Raw benchmark valueCI 1,437 - 1,461

Text Arena · Industry Writing And Literature And Language

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #92

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,381
Percentile: 71.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_writing_and_literature_and_language. Source rank: #114. Votes: 1547. Organization: tencent. License: tencent-hunyuan-community.

71.9% percentile inside its fair comparison set

1,381Raw benchmark valueCI 1,366 - 1,397

Text Arena · Expert · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena expert leaderboard.

Rank #72

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,422
Percentile: 74.2%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: expert. Source rank: #88. Votes: 652. Organization: tencent. License: tencent-hunyuan-community.

74.2% percentile inside its fair comparison set

1,422Raw benchmark valueCI 1,399 - 1,445

Text Arena · Industry Business And Management And Financial Operations · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_business_and_management_and_financial_operations leaderboard.

Rank #79

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,401
Percentile: 75.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_business_and_management_and_financial_operations. Source rank: #95. Votes: 1374. Organization: tencent. License: tencent-hunyuan-community.

75.5% percentile inside its fair comparison set

1,401Raw benchmark valueCI 1,385 - 1,418

Text Arena · Industry Entertainment And Sports And Media · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_entertainment_and_sports_and_media leaderboard.

Rank #98

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,357
Percentile: 70%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_entertainment_and_sports_and_media. Source rank: #118. Votes: 1318. Organization: tencent. License: tencent-hunyuan-community.

70% percentile inside its fair comparison set

1,357Raw benchmark valueCI 1,340 - 1,374

Text Arena · Industry Legal And Government · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_legal_and_government leaderboard.

Rank #64

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,429
Percentile: 78.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_legal_and_government. Source rank: #77. Votes: 527. Organization: tencent. License: tencent-hunyuan-community.

78.9% percentile inside its fair comparison set

1,429Raw benchmark valueCI 1,403 - 1,455

Text Arena · Industry Life And Physical And Social Science · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_life_and_physical_and_social_science leaderboard.

Rank #86

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,421
Percentile: 73.7%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_life_and_physical_and_social_science. Source rank: #101. Votes: 1223. Organization: tencent. License: tencent-hunyuan-community.

73.7% percentile inside its fair comparison set

1,421Raw benchmark valueCI 1,404 - 1,438

Text Arena · Industry Mathematical · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_mathematical leaderboard.

Rank #46

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,440
Percentile: 85.4%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_mathematical. Source rank: #55. Votes: 423. Organization: tencent. License: tencent-hunyuan-community.

85.4% percentile inside its fair comparison set

1,440Raw benchmark valueCI 1,412 - 1,468

Text Arena · Industry Medicine And Healthcare · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_medicine_and_healthcare leaderboard.

Rank #92

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,410
Percentile: 69.2%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_medicine_and_healthcare. Source rank: #111. Votes: 468. Organization: tencent. License: tencent-hunyuan-community.

69.2% percentile inside its fair comparison set

1,410Raw benchmark valueCI 1,382 - 1,438

Text Arena · Industry Software And It Services · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_software_and_it_services leaderboard.

Rank #86

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,426
Percentile: 73.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_software_and_it_services. Source rank: #103. Votes: 2695. Organization: tencent. License: tencent-hunyuan-community.

73.8% percentile inside its fair comparison set

1,426Raw benchmark valueCI 1,415 - 1,438

Text Arena · Industry Writing And Literature And Language · No Style Control

AR · Professional reasoning · Human

Observed user preference in Arena's Text Arena industry_writing_and_literature_and_language leaderboard.

Rank #90

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,374
Percentile: 72.5%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: industry_writing_and_literature_and_language. Source rank: #112. Votes: 1547. Organization: tencent. License: tencent-hunyuan-community.

72.5% percentile inside its fair comparison set

1,374Raw benchmark valueCI 1,359 - 1,390

Multilingual8 benchmarks70.5%

Text Arena · Chinese

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #60

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,468
Percentile: 80%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: chinese. Source rank: #74. Votes: 411. Organization: tencent. License: tencent-hunyuan-community.

80% percentile inside its fair comparison set

1,468Raw benchmark valueCI 1,438 - 1,498

Text Arena · French

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #70

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,440
Percentile: 68.1%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: french. Source rank: #87. Votes: 234. Organization: tencent. License: tencent-hunyuan-community.

68.1% percentile inside its fair comparison set

1,440Raw benchmark valueCI 1,400 - 1,480

Text Arena · Russian

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #62

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,415
Percentile: 78.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: russian. Source rank: #81. Votes: 674. Organization: tencent. License: tencent-hunyuan-community.

78.9% percentile inside its fair comparison set

1,415Raw benchmark valueCI 1,392 - 1,439

Text Arena · Spanish

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #87

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 59.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: spanish. Source rank: #108. Votes: 202. Organization: tencent. License: tencent-hunyuan-community.

59.8% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,350 - 1,433

Text Arena · Chinese · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena chinese leaderboard.

Rank #81

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,446
Percentile: 72.9%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: chinese. Source rank: #97. Votes: 411. Organization: tencent. License: tencent-hunyuan-community.

72.9% percentile inside its fair comparison set

1,446Raw benchmark valueCI 1,415 - 1,476

Text Arena · French · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena french leaderboard.

Rank #71

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,425
Percentile: 67.6%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: french. Source rank: #87. Votes: 234. Organization: tencent. License: tencent-hunyuan-community.

67.6% percentile inside its fair comparison set

1,425Raw benchmark valueCI 1,385 - 1,466

Text Arena · Russian · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena russian leaderboard.

Rank #68

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,407
Percentile: 76.8%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: russian. Source rank: #85. Votes: 674. Organization: tencent. License: tencent-hunyuan-community.

76.8% percentile inside its fair comparison set

1,407Raw benchmark valueCI 1,383 - 1,430

Text Arena · Spanish · No Style Control

AR · Multilingual · Human

Observed user preference in Arena's Text Arena spanish leaderboard.

Rank #86

verified runtimeexact aliasBackground only

Raw row drilldownsource row, percentile, last updated, eligibility

Source: Arena
Raw value: 1,392
Percentile: 60.3%
Last updated: recent
Eligibility: preview_model

Parsed from Arena leaderboard dataset row `hunyuan-hy3-preview`. Category: spanish. Source rank: #104. Votes: 202. Organization: tencent. License: tencent-hunyuan-community.

60.3% percentile inside its fair comparison set

1,392Raw benchmark valueCI 1,351 - 1,434

Source links and registry checks

official

Arena

Jun 20, 2026

hunyuan-hy3-preview

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks

Loading model evidence.

hunyuan-hy3-preview

Current snapshot.

Source-linked scores by benchmark

Source links and registry checks