GeekRadar / The comparison desk
COMPARE.
AI models
alibaba

Qwen3-Next 80B-A3B Instruct

AI model

Context window
262,144 tokens
Maximum output
32,768 tokens
Open weights
Yes

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Qwen3-Next 80B-A3B Instruct 1417.0 Qwen3-Next-80B-A3B-Instruct · 95% interval 1412.3–1421.7 · 22,753 votes · 2026-10-28

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,459.76 rating

Confidence interval: 1,445.54 – 1,473.99

1729 votes

Category
chinese
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,440.54 rating

Confidence interval: 1,431.96 – 1,449.13

4811 votes

Category
coding
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,333.36 rating

Confidence interval: 1,322.59 – 1,344.12

3035 votes

Category
creative_writing
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,421.31 rating

Confidence interval: 1,414.92 – 1,427.69

9736 votes

Category
english
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,404.28 rating

Confidence interval: 1,397.75 – 1,410.81

16043 votes

Category
exclude_ties
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,416.32 rating

Confidence interval: 1,398.76 – 1,433.87

1098 votes

Category
expert
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,413.88 rating

Confidence interval: 1,383.60 – 1,444.15

388 votes

Category
french
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,416.83 rating

Confidence interval: 1,390.65 – 1,443.00

491 votes

Category
german
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,427.46 rating

Confidence interval: 1,421.48 – 1,433.45

11997 votes

Category
hard_prompts
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,431.06 rating

Confidence interval: 1,422.78 – 1,439.33

5217 votes

Category
hard_prompts_english
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,421.74 rating

Confidence interval: 1,412.52 – 1,430.96

4286 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,348.85 rating

Confidence interval: 1,339.79 – 1,357.91

4228 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,431.20 rating

Confidence interval: 1,415.61 – 1,446.79

1414 votes

Category
industry_legal_and_government
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,434.19 rating

Confidence interval: 1,424.22 – 1,444.16

3528 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,433.80 rating

Confidence interval: 1,416.13 – 1,451.47

1077 votes

Category
industry_mathematical
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,455.13 rating

Confidence interval: 1,437.96 – 1,472.29

1180 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,446.98 rating

Confidence interval: 1,440.08 – 1,453.88

8045 votes

Category
industry_software_and_it_services
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,354.81 rating

Confidence interval: 1,346.41 – 1,363.20

5064 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,388.45 rating

Confidence interval: 1,380.93 – 1,395.97

6416 votes

Category
instruction_following
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,395.47 rating

Confidence interval: 1,362.57 – 1,428.37

319 votes

Category
japanese
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,363.95 rating

Confidence interval: 1,337.35 – 1,390.55

525 votes

Category
korean
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,402.44 rating

Confidence interval: 1,394.23 – 1,410.65

5395 votes

Category
longer_query
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,439.91 rating

Confidence interval: 1,423.19 – 1,456.63

1203 votes

Category
math
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,415.57 rating

Confidence interval: 1,406.13 – 1,425.00

4012 votes

Category
multi_turn
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,406.87 rating

Confidence interval: 1,401.03 – 1,412.72

12983 votes

Category
non_english
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,417.01 rating

Confidence interval: 1,412.28 – 1,421.75

22753 votes

Category
overall
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,398.06 rating

Confidence interval: 1,379.54 – 1,416.58

1022 votes

Category
polish
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,403.52 rating

Confidence interval: 1,389.20 – 1,417.85

1611 votes

Category
russian
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Qwen3-Next 80B-A3B Instruct

1,434.65 rating

Confidence interval: 1,411.65 – 1,457.65

688 votes

Category
spanish
Evaluated configuration
Qwen3-Next-80B-A3B-Instruct
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,460.09 rating

Confidence interval: 1,445.87 – 1,474.31

1729 votes

Category
chinese
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,440.48 rating

Confidence interval: 1,431.90 – 1,449.06

4817 votes

Category
coding
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,333.37 rating

Confidence interval: 1,322.61 – 1,344.13

3033 votes

Category
creative_writing
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,421.33 rating

Confidence interval: 1,414.94 – 1,427.71

9735 votes

Category
english
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,404.29 rating

Confidence interval: 1,397.76 – 1,410.82

16046 votes

Category
exclude_ties
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,416.38 rating

Confidence interval: 1,398.80 – 1,433.95

1096 votes

Category
expert
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,413.41 rating

Confidence interval: 1,383.13 – 1,443.70

388 votes

Category
french
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,416.29 rating

Confidence interval: 1,390.15 – 1,442.43

492 votes

Category
german
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,427.48 rating

Confidence interval: 1,421.49 – 1,433.46

11995 votes

Category
hard_prompts
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,430.90 rating

Confidence interval: 1,422.62 – 1,439.17

5216 votes

Category
hard_prompts_english
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,421.88 rating

Confidence interval: 1,412.66 – 1,431.10

4287 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,348.95 rating

Confidence interval: 1,339.88 – 1,358.01

4225 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,430.99 rating

Confidence interval: 1,415.40 – 1,446.58

1414 votes

Category
industry_legal_and_government
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,434.48 rating

Confidence interval: 1,424.52 – 1,444.44

3531 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,433.83 rating

Confidence interval: 1,416.17 – 1,451.50

1077 votes

Category
industry_mathematical
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,455.41 rating

Confidence interval: 1,438.26 – 1,472.56

1182 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,446.99 rating

Confidence interval: 1,440.09 – 1,453.89

8050 votes

Category
industry_software_and_it_services
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,354.81 rating

Confidence interval: 1,346.41 – 1,363.20

5063 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,388.34 rating

Confidence interval: 1,380.82 – 1,395.85

6412 votes

Category
instruction_following
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,395.28 rating

Confidence interval: 1,362.28 – 1,428.28

318 votes

Category
japanese
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,363.36 rating

Confidence interval: 1,336.70 – 1,390.02

523 votes

Category
korean
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,402.53 rating

Confidence interval: 1,394.33 – 1,410.74

5396 votes

Category
longer_query
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,440.05 rating

Confidence interval: 1,423.33 – 1,456.77

1204 votes

Category
math
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,415.80 rating

Confidence interval: 1,406.36 – 1,425.23

4015 votes

Category
multi_turn
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,406.85 rating

Confidence interval: 1,401.00 – 1,412.69

12991 votes

Category
non_english
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,417.02 rating

Confidence interval: 1,412.29 – 1,421.76

22760 votes

Category
overall
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,398.09 rating

Confidence interval: 1,379.57 – 1,416.60

1022 votes

Category
polish
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,403.86 rating

Confidence interval: 1,389.58 – 1,418.15

1618 votes

Category
russian
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

Qwen3-Next 80B-A3B Instruct

1,435.04 rating

Confidence interval: 1,412.01 – 1,458.07

687 votes

Category
spanish
Evaluated configuration
qwen3-next-80b-a3b-instruct
Evaluation type
human preference

Run it locally

Inference · memory estimate

Qwen3-Next 80B-A3B Instruct

151.5 GiB checkpoint tensor payload · 81.32 B stored tensor elements · BF16

Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Qwen3-Next 80B-A3B Instruct

Qwen3-Next 80B-A3B Instruct / Pricing tiers 1
Provider / scopeRateCurrencyAmount / unit
DeepInfraQwen/Qwen3-Next-80B-A3B-InstructStandard InputUSD0.09 / 1M tokens
OutputUSD1.1 / 1M tokens

Capabilities

Context window
262,144 tokens
Maximum output
32,768 tokens
Reasoning
No
Tool calling
Yes
Open weights
Yes
Knowledge cutoff
2025-04
Checkpoint tensor payload
151.48 GiB

Input modalities

  • text

Output modalities

  • text

Processing & platform

Architecture
qwen3_next