Qwen3.8 27B
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.
GeekRadar / The comparison desk
COMPARE.
AI modelsQwen3.8 27B
AI model Released
- 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.
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
Qwen3.8 27B
1,500.99 rating
Confidence interval: 1,487.83 – 1,514.14
2120 votes
- Category
- chinese
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,481.37 rating
Confidence interval: 1,473.79 – 1,488.96
7536 votes
- Category
- coding
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,379.11 rating
Confidence interval: 1,370.19 – 1,388.03
5601 votes
- Category
- creative_writing
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,446.97 rating
Confidence interval: 1,440.29 – 1,453.65
10380 votes
- Category
- english
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,431.83 rating
Confidence interval: 1,425.47 – 1,438.19
20328 votes
- Category
- exclude_ties
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,481.51 rating
Confidence interval: 1,470.44 – 1,492.58
3024 votes
- Category
- expert
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,464.72 rating
Confidence interval: 1,444.65 – 1,484.79
974 votes
- Category
- french
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,444.09 rating
Confidence interval: 1,416.39 – 1,471.80
458 votes
- Category
- german
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,458.09 rating
Confidence interval: 1,452.45 – 1,463.73
18028 votes
- Category
- hard_prompts
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,460.46 rating
Confidence interval: 1,452.63 – 1,468.28
6879 votes
- Category
- hard_prompts_english
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,439.15 rating
Confidence interval: 1,430.31 – 1,447.98
5050 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,390.10 rating
Confidence interval: 1,381.96 – 1,398.25
6851 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,446.75 rating
Confidence interval: 1,433.46 – 1,460.03
2148 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,467.64 rating
Confidence interval: 1,458.24 – 1,477.05
4445 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,459.98 rating
Confidence interval: 1,445.36 – 1,474.61
1641 votes
- Category
- industry_mathematical
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,441.71 rating
Confidence interval: 1,427.19 – 1,456.23
1807 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,476.69 rating
Confidence interval: 1,470.06 – 1,483.32
10781 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,401.44 rating
Confidence interval: 1,393.55 – 1,409.34
7307 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,434.71 rating
Confidence interval: 1,427.79 – 1,441.63
9844 votes
- Category
- instruction_following
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,395.55 rating
Confidence interval: 1,368.24 – 1,422.86
522 votes
- Category
- japanese
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,379.82 rating
Confidence interval: 1,351.87 – 1,407.77
459 votes
- Category
- korean
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,447.68 rating
Confidence interval: 1,441.07 – 1,454.29
12166 votes
- Category
- longer_query
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,455.86 rating
Confidence interval: 1,440.06 – 1,471.67
1400 votes
- Category
- math
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,438.22 rating
Confidence interval: 1,428.66 – 1,447.78
4334 votes
- Category
- multi_turn
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,426.38 rating
Confidence interval: 1,420.69 – 1,432.07
17393 votes
- Category
- non_english
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,438.44 rating
Confidence interval: 1,433.61 – 1,443.28
27779 votes
- Category
- overall
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,446.45 rating
Confidence interval: 1,418.45 – 1,474.45
435 votes
- Category
- polish
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,411.32 rating
Confidence interval: 1,401.04 – 1,421.61
3656 votes
- Category
- russian
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
Qwen3.8 27B
1,446.03 rating
Confidence interval: 1,426.05 – 1,466.01
913 votes
- Category
- spanish
- Evaluated configuration
- Qwen3.8-27B
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,502.28 rating
Confidence interval: 1,487.87 – 1,516.69
1743 votes
- Category
- chinese
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,482.71 rating
Confidence interval: 1,474.72 – 1,490.70
6417 votes
- Category
- coding
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,384.23 rating
Confidence interval: 1,374.71 – 1,393.75
4668 votes
- Category
- creative_writing
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,447.48 rating
Confidence interval: 1,440.49 – 1,454.47
8818 votes
- Category
- english
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,434.63 rating
Confidence interval: 1,427.95 – 1,441.30
17023 votes
- Category
- exclude_ties
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,480.60 rating
Confidence interval: 1,468.73 – 1,492.46
2583 votes
- Category
- expert
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,468.19 rating
Confidence interval: 1,446.25 – 1,490.13
791 votes
- Category
- french
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,436.29 rating
Confidence interval: 1,405.80 – 1,466.79
368 votes
- Category
- german
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,459.13 rating
Confidence interval: 1,453.21 – 1,465.06
15193 votes
- Category
- hard_prompts
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,460.18 rating
Confidence interval: 1,451.92 – 1,468.45
5852 votes
- Category
- hard_prompts_english
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,434.99 rating
Confidence interval: 1,425.56 – 1,444.41
4265 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,394.64 rating
Confidence interval: 1,386.02 – 1,403.26
5740 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,446.70 rating
Confidence interval: 1,432.39 – 1,461.01
1799 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,467.68 rating
Confidence interval: 1,457.55 – 1,477.81
3689 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,461.66 rating
Confidence interval: 1,445.54 – 1,477.78
1331 votes
- Category
- industry_mathematical
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,440.79 rating
Confidence interval: 1,425.05 – 1,456.52
1511 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,477.82 rating
Confidence interval: 1,470.85 – 1,484.79
9127 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,406.38 rating
Confidence interval: 1,397.98 – 1,414.77
6078 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,436.80 rating
Confidence interval: 1,429.51 – 1,444.10
8237 votes
- Category
- instruction_following
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,392.29 rating
Confidence interval: 1,361.50 – 1,423.09
414 votes
- Category
- japanese
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,387.07 rating
Confidence interval: 1,357.08 – 1,417.06
394 votes
- Category
- korean
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,449.25 rating
Confidence interval: 1,442.33 – 1,456.18
10316 votes
- Category
- longer_query
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,456.79 rating
Confidence interval: 1,439.64 – 1,473.93
1171 votes
- Category
- math
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,441.56 rating
Confidence interval: 1,431.42 – 1,451.71
3711 votes
- Category
- multi_turn
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,429.40 rating
Confidence interval: 1,423.40 – 1,435.40
14467 votes
- Category
- non_english
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,440.55 rating
Confidence interval: 1,435.48 – 1,445.62
23294 votes
- Category
- overall
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,441.74 rating
Confidence interval: 1,411.24 – 1,472.25
364 votes
- Category
- polish
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,416.01 rating
Confidence interval: 1,404.86 – 1,427.15
2983 votes
- Category
- russian
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
Qwen3.8 27B
1,443.37 rating
Confidence interval: 1,421.44 – 1,465.30
746 votes
- Category
- spanish
- Evaluated configuration
- qwen3.8-27b
- Evaluation type
- human preference
Run it locally
API cost
Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.
Qwen3.8 27B
Qwen3.8 27B / Pricing tiers
| Provider / scope | Rate | Currency | Amount / unit |
|---|---|---|---|
| DeepInfraQwen/Qwen3.8-27BStandard | Cached input | USD | 0.05 / 1M tokens |
| Input | USD | 0.2 / 1M tokens | |
| Output | USD | 2.5 / 1M tokens |
Capabilities
- Context window
- 262,144 tokens
- Maximum output
- 32,768 tokens
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Checkpoint tensor payload
- 51.75 GiB
Input modalities
- text
- image
- video
Output modalities
- text
Processing & platform
- Architecture
- qwen3_5
0 selected Compare 0