{"fetched_at":"2026-08-13T15:08:33.755160+00:00","cached_at":"2026-07-29T23:09:40.257102+00:00","cache_expires_at":"2026-07-30T05:09:40.257102+00:00","from_cache":true,"name_full_model":"llama3","namespace":"library","model_base_name":"llama3","active_tag_part":"8b","active_tag_full_name":"llama3:8b","source_url":"https://ollama.com/library/llama3:8b","summary":"Meta Llama 3: The most capable openly available LLM to date","pull_count_str":"0","pull_count":0,"last_updated_str":"","last_updated_iso":"2026-08-13T15:08:34.255049+00:00","capabilities":[],"sizes":[],"readme_content":"
\n

Llama 3

\n

The most capable openly available LLM to date.

\n

\n

Meta Llama 3, a family of models developed by Meta Inc. are new state-of-the-art , available in both 8B and 70B parameter sizes (pre-trained or instruction-tuned).

\n

Llama 3 instruction-tuned models are fine-tuned and optimized for dialogue/chat use cases and outperform many of the available open-source chat models on common benchmarks.

\n

\n

\n

CLI

\n

Open the terminal and run ollama run llama3

\n

API

\n

Example using curl:

\n
curl -X POST http://localhost:11434/api/generate -d '{\n  \"model\": \"llama3\",\n  \"prompt\":\"Why is the sky blue?\"\n }'\n
\n

API documentation

\n

Model variants

\n

Instruct is fine-tuned for chat/dialogue use cases.

\n

Example:\nollama run llama3\nollama run llama3:70b

\n

Pre-trained is the base model.

\n

Example:\nollama run llama3:text\nollama run llama3:70b-text

\n

References

\n

Introducing Meta Llama 3: The most capable openly available LLM to date

\n
","tag_command":null,"tag_files_summary":[],"all_tags_dropdown_summary":[],"all_tags_page_url":"https://ollama.com/library/llama3/tags","total_tags_count_from_link":0}