Llama 3 instruction-tuned architecture
The captured configuration identifies LlamaForCausalLM with model type llama and Transformers support.
Open Source Model Profile · meta-llama
Meta-Llama-3-8B-Instruct
Meta-Llama-3-8B-Instruct is an 8.03B-parameter Llama 3 instruction-tuned model from Meta. According to the model card, it is optimized for dialogue use.
- Publisher
- meta-llama
- Task
- text-generation
- Model type
- llama
- License
- llama3
- Library
- transformers
- Publication status
- Approved for indexing
Model overview
Meta-Llama-3-8B-Instruct is published by Meta as an instruction-tuned member of the Llama 3 family. The captured configuration identifies LlamaForCausalLM, and Safetensors metadata reports about 8.03B parameters under llama3. According to the model card, the family spans 8B and 70B pretrained and instruction-tuned variants.
Recorded capabilities
Dialogue optimization
According to the model card, the instruction-tuned models target dialogue use and combine supervised finetuning with reinforcement learning from human feedback.
Documented 8k context
According to the model card, the 8B model uses 8k context with grouped-query attention.
Documented training footprint
According to the model card, Llama 3 pretraining used Meta compute clusters, with the 8B model accounting for 1.3M GPU hours in the reported footprint.
Transformers inference path
According to the model card, conversational inference can run through the Transformers pipeline or Auto classes with generate().
Use cases in the source record
- Dialogue and conversational text-generation workflows using the documented Transformers inference path.
- Text and code generation tasks within the scope of the card's text-in, text-and-code-out description.
Limitations and unknowns
- The card contains extensive publisher-reported benchmark tables; those scores are publisher claims and were not independently measured by Ethen.
- Provider state is historical snapshot data, including one live and one error entry, and should be refreshed before being presented as current.
- According to the model card, testing was conducted in English and outputs may be inaccurate or biased, so application-specific safety testing is advised.
Source and provenance
Source: meta-llama/Meta-Llama-3-8B-Instruct
Captured: Unknown. Processed: 2026-09-07T19:34:51.900337+00:00.
Model Details Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety. Model developers Meta Variations Llama 3 comes in two sizes — 8B and 70B parameters — in pre-trained and instruction tuned variants. Input Models input text only. Output Models generate text…
F001F002F003F004F005F006F007F008F009F010F011F012F014F015F016F017F022F023F024F025F026F030