Model
Inference & ServingCoding
Code Llama
1Meta's 2Llama 32 4fine-tuned 5for 6coding
the six Ws · specification
W1
Who
Meta AI released Code Llama, a code-specialized fine-tune of Llama 2.
W2
What
A family of 7B to 70B models for code generation, infilling, and instruction-following on programming tasks.
W3
Where
Released on Hugging Face and GitHub under the Meta and facebookresearch organizations.
W4
When
Code Llama launched in August 2023, with a 70B version added in January 2024.
W5
Why
It was one of the first widely-adopted open code models, powering many downstream coding assistants.
W6
With
Runs via transformers, llama.cpp, or vLLM, and is available through Ollama for local use.
W7
Watch
Open-weight release by Meta under the Llama 2 Community License, which restricts use above 700 million monthly active users.
alternatives
works with