Llama-3.2-3B-Instruct
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 modelsLlama-3.2-3B-Instruct
AI model Released
- Context window
- 131,072 tokens
- Maximum output
- 131,072 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
Llama-3.2-3B-Instruct
1,017.21 rating
Confidence interval: 992.83 – 1,041.58
721 votes
- Category
- chinese
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,097.43 rating
Confidence interval: 1,081.62 – 1,113.23
1351 votes
- Category
- coding
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,093.87 rating
Confidence interval: 1,075.00 – 1,112.74
1070 votes
- Category
- creative_writing
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,179.60 rating
Confidence interval: 1,170.00 – 1,189.21
4157 votes
- Category
- english
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
933.13 rating
Confidence interval: 919.79 – 946.48
5442 votes
- Category
- exclude_ties
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,089.91 rating
Confidence interval: 1,065.73 – 1,114.10
499 votes
- Category
- expert
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,056.13 rating
Confidence interval: 1,015.67 – 1,096.58
232 votes
- Category
- german
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,094.51 rating
Confidence interval: 1,082.37 – 1,106.65
2216 votes
- Category
- hard_prompts
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,140.80 rating
Confidence interval: 1,125.85 – 1,155.75
1377 votes
- Category
- hard_prompts_english
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,073.60 rating
Confidence interval: 1,053.98 – 1,093.22
1012 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,083.34 rating
Confidence interval: 1,065.77 – 1,100.90
1149 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,115.29 rating
Confidence interval: 1,087.35 – 1,143.23
506 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,105.81 rating
Confidence interval: 1,088.49 – 1,123.14
1355 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,128.43 rating
Confidence interval: 1,111.70 – 1,145.16
980 votes
- Category
- industry_mathematical
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,104.20 rating
Confidence interval: 1,074.20 – 1,134.19
441 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,103.17 rating
Confidence interval: 1,090.28 – 1,116.05
2165 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,095.50 rating
Confidence interval: 1,081.73 – 1,109.28
2118 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,088.74 rating
Confidence interval: 1,078.00 – 1,099.48
3171 votes
- Category
- instruction_following
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,100.15 rating
Confidence interval: 1,082.40 – 1,117.91
1046 votes
- Category
- longer_query
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,125.88 rating
Confidence interval: 1,110.20 – 1,141.57
1136 votes
- Category
- math
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,105.08 rating
Confidence interval: 1,088.10 – 1,122.06
1299 votes
- Category
- multi_turn
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,018.70 rating
Confidence interval: 1,007.63 – 1,029.77
3779 votes
- Category
- non_english
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
1,109.51 rating
Confidence interval: 1,101.96 – 1,117.06
7936 votes
- Category
- overall
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
Llama-3.2-3B-Instruct
948.73 rating
Confidence interval: 926.31 – 971.16
933 votes
- Category
- russian
- Evaluated configuration
- Llama 3.2 3B Instruct
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,017.25 rating
Confidence interval: 992.87 – 1,041.64
721 votes
- Category
- chinese
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,097.53 rating
Confidence interval: 1,081.72 – 1,113.33
1351 votes
- Category
- coding
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,094.03 rating
Confidence interval: 1,075.15 – 1,112.90
1070 votes
- Category
- creative_writing
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,179.55 rating
Confidence interval: 1,169.95 – 1,189.16
4157 votes
- Category
- english
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
933.14 rating
Confidence interval: 919.80 – 946.49
5442 votes
- Category
- exclude_ties
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,089.95 rating
Confidence interval: 1,065.76 – 1,114.14
499 votes
- Category
- expert
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,056.12 rating
Confidence interval: 1,015.69 – 1,096.56
232 votes
- Category
- german
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,094.50 rating
Confidence interval: 1,082.36 – 1,106.64
2216 votes
- Category
- hard_prompts
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,140.69 rating
Confidence interval: 1,125.73 – 1,155.64
1377 votes
- Category
- hard_prompts_english
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,073.64 rating
Confidence interval: 1,054.02 – 1,093.26
1012 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,083.40 rating
Confidence interval: 1,065.83 – 1,100.97
1149 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,115.04 rating
Confidence interval: 1,087.09 – 1,142.98
506 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,105.79 rating
Confidence interval: 1,088.46 – 1,123.12
1355 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,128.43 rating
Confidence interval: 1,111.69 – 1,145.16
980 votes
- Category
- industry_mathematical
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,104.25 rating
Confidence interval: 1,074.26 – 1,134.25
441 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,103.25 rating
Confidence interval: 1,090.36 – 1,116.14
2165 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,095.57 rating
Confidence interval: 1,081.79 – 1,109.35
2118 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,088.75 rating
Confidence interval: 1,078.01 – 1,099.49
3171 votes
- Category
- instruction_following
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,100.20 rating
Confidence interval: 1,082.44 – 1,117.95
1046 votes
- Category
- longer_query
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,125.85 rating
Confidence interval: 1,110.16 – 1,141.53
1136 votes
- Category
- math
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,105.08 rating
Confidence interval: 1,088.10 – 1,122.06
1299 votes
- Category
- multi_turn
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,018.79 rating
Confidence interval: 1,007.72 – 1,029.86
3779 votes
- Category
- non_english
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
1,109.50 rating
Confidence interval: 1,101.94 – 1,117.05
7936 votes
- Category
- overall
- Evaluated configuration
- llama-3.2-3b-instruct
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Llama-3.2-3B-Instruct
948.89 rating
Confidence interval: 926.46 – 971.32
933 votes
- Category
- russian
- Evaluated configuration
- llama-3.2-3b-instruct
- 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.
Llama-3.2-3B-Instruct
Capabilities
- Context window
- 131,072 tokens
- Maximum output
- 131,072 tokens
- Reasoning
- No
- Tool calling
- Yes
- Open weights
- Yes
- Knowledge cutoff
- 2023-12
- Model license
- llama3.2
- Checkpoint tensor payload
- 5.98 GiB
Input modalities
- text
Output modalities
- text
0 selected Compare 0