Llama-3.1-70B-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-3.1-70B-Instruct
AI model Released
- Context window
- 128,000 tokens
- Maximum output
- 4,096 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-3.1-70B-Instruct
1,215.20 rating
Confidence interval: 1,205.33 – 1,225.06
4846 votes
- Category
- chinese
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,259.39 rating
Confidence interval: 1,252.30 – 1,266.49
9389 votes
- Category
- coding
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,231.57 rating
Confidence interval: 1,223.81 – 1,239.33
8250 votes
- Category
- creative_writing
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,293.63 rating
Confidence interval: 1,288.81 – 1,298.45
29694 votes
- Category
- english
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,176.52 rating
Confidence interval: 1,170.67 – 1,182.37
34840 votes
- Category
- exclude_ties
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,208.45 rating
Confidence interval: 1,196.96 – 1,219.94
2924 votes
- Category
- expert
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,261.61 rating
Confidence interval: 1,232.95 – 1,290.28
501 votes
- Category
- french
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,221.67 rating
Confidence interval: 1,204.62 – 1,238.73
1353 votes
- Category
- german
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,241.08 rating
Confidence interval: 1,235.27 – 1,246.90
14981 votes
- Category
- hard_prompts
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,261.21 rating
Confidence interval: 1,254.11 – 1,268.32
9133 votes
- Category
- hard_prompts_english
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,235.01 rating
Confidence interval: 1,226.51 – 1,243.50
6381 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,229.77 rating
Confidence interval: 1,222.44 – 1,237.10
8719 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,283.54 rating
Confidence interval: 1,271.76 – 1,295.31
3177 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,275.99 rating
Confidence interval: 1,268.61 – 1,283.36
9381 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,253.44 rating
Confidence interval: 1,245.32 – 1,261.56
6524 votes
- Category
- industry_mathematical
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,257.06 rating
Confidence interval: 1,244.51 – 1,269.61
2821 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,259.75 rating
Confidence interval: 1,253.67 – 1,265.82
14811 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,235.64 rating
Confidence interval: 1,229.42 – 1,241.87
14962 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,231.07 rating
Confidence interval: 1,225.79 – 1,236.36
21910 votes
- Category
- instruction_following
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,132.50 rating
Confidence interval: 1,115.20 – 1,149.79
1428 votes
- Category
- japanese
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,140.41 rating
Confidence interval: 1,117.37 – 1,163.46
873 votes
- Category
- korean
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,240.63 rating
Confidence interval: 1,233.03 – 1,248.23
7622 votes
- Category
- longer_query
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,251.64 rating
Confidence interval: 1,244.02 – 1,259.26
7677 votes
- Category
- math
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,256.13 rating
Confidence interval: 1,248.93 – 1,263.33
10405 votes
- Category
- multi_turn
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,219.13 rating
Confidence interval: 1,214.20 – 1,224.07
25546 votes
- Category
- non_english
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,260.88 rating
Confidence interval: 1,257.17 – 1,264.60
55240 votes
- Category
- overall
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,233.49 rating
Confidence interval: 1,225.42 – 1,241.56
7285 votes
- Category
- russian
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
Llama-3.1-70B-Instruct
1,253.11 rating
Confidence interval: 1,227.90 – 1,278.31
606 votes
- Category
- spanish
- Evaluated configuration
- Llama 3.1 70B Instruct
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,215.24 rating
Confidence interval: 1,205.38 – 1,225.11
4846 votes
- Category
- chinese
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,259.49 rating
Confidence interval: 1,252.40 – 1,266.59
9389 votes
- Category
- coding
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,231.73 rating
Confidence interval: 1,223.96 – 1,239.49
8250 votes
- Category
- creative_writing
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,293.58 rating
Confidence interval: 1,288.76 – 1,298.40
29694 votes
- Category
- english
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,176.53 rating
Confidence interval: 1,170.68 – 1,182.38
34840 votes
- Category
- exclude_ties
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,208.49 rating
Confidence interval: 1,197.00 – 1,219.99
2924 votes
- Category
- expert
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,261.30 rating
Confidence interval: 1,232.62 – 1,289.97
501 votes
- Category
- french
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,221.67 rating
Confidence interval: 1,204.67 – 1,238.67
1353 votes
- Category
- german
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,241.07 rating
Confidence interval: 1,235.26 – 1,246.89
14981 votes
- Category
- hard_prompts
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,261.10 rating
Confidence interval: 1,253.99 – 1,268.20
9133 votes
- Category
- hard_prompts_english
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,235.04 rating
Confidence interval: 1,226.55 – 1,243.54
6381 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,229.83 rating
Confidence interval: 1,222.49 – 1,237.16
8719 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,283.28 rating
Confidence interval: 1,271.50 – 1,295.06
3177 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,275.96 rating
Confidence interval: 1,268.58 – 1,283.34
9381 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,253.43 rating
Confidence interval: 1,245.32 – 1,261.55
6524 votes
- Category
- industry_mathematical
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,257.12 rating
Confidence interval: 1,244.56 – 1,269.67
2821 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,259.83 rating
Confidence interval: 1,253.75 – 1,265.91
14811 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,235.71 rating
Confidence interval: 1,229.48 – 1,241.93
14962 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,231.08 rating
Confidence interval: 1,225.80 – 1,236.37
21910 votes
- Category
- instruction_following
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,132.33 rating
Confidence interval: 1,115.11 – 1,149.54
1428 votes
- Category
- japanese
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,140.46 rating
Confidence interval: 1,117.43 – 1,163.48
873 votes
- Category
- korean
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,240.67 rating
Confidence interval: 1,233.07 – 1,248.27
7622 votes
- Category
- longer_query
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,251.60 rating
Confidence interval: 1,243.98 – 1,259.21
7677 votes
- Category
- math
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,256.13 rating
Confidence interval: 1,248.93 – 1,263.33
10405 votes
- Category
- multi_turn
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,219.23 rating
Confidence interval: 1,214.29 – 1,224.16
25546 votes
- Category
- non_english
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,260.87 rating
Confidence interval: 1,257.16 – 1,264.58
55240 votes
- Category
- overall
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,233.64 rating
Confidence interval: 1,225.57 – 1,241.72
7285 votes
- Category
- russian
- Evaluated configuration
- llama-3.1-70b-instruct
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
Llama-3.1-70B-Instruct
1,253.03 rating
Confidence interval: 1,227.80 – 1,278.26
606 votes
- Category
- spanish
- Evaluated configuration
- llama-3.1-70b-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-3.1-70B-Instruct
Capabilities
- Context window
- 128,000 tokens
- Maximum output
- 4,096 tokens
- Reasoning
- No
- Tool calling
- Yes
- Open weights
- Yes
- Knowledge cutoff
- 2023-12
- Checkpoint tensor payload
- 131.42 GiB
Input modalities
- text
Output modalities
- text
0 selected Compare 0