codellama-70b-instruct
Not enough verified architecture information for a reliable VRAM estimate.
GeekRadar / The comparison desk
COMPARE.
AI modelscodellama-70b-instruct
Model evaluation
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 / englishhuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,095.82 rating
Confidence interval: 1,073.96 – 1,117.67
695 votes
- Category
- english
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
codellama-70b-instruct
872.25 rating
Confidence interval: 844.91 – 899.59
809 votes
- Category
- exclude_ties
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,052.01 rating
Confidence interval: 1,021.51 – 1,082.52
306 votes
- Category
- hard_prompts
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,077.38 rating
Confidence interval: 1,041.24 – 1,113.52
207 votes
- Category
- hard_prompts_english
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,057.58 rating
Confidence interval: 1,026.06 – 1,089.11
325 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,014.77 rating
Confidence interval: 978.28 – 1,051.27
262 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,023.55 rating
Confidence interval: 993.97 – 1,053.14
358 votes
- Category
- instruction_following
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
codellama-70b-instruct
992.28 rating
Confidence interval: 964.43 – 1,020.13
448 votes
- Category
- non_english
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
codellama-70b-instruct
1,057.20 rating
Confidence interval: 1,039.22 – 1,075.18
1143 votes
- Category
- overall
- Evaluated configuration
- codellama-70b-instruct
- Evaluation type
- human preference
Run it locally
Capabilities
- Model license
- Llama 2 Community
0 selected Compare 0