Because the runtime includes llama.cpp natively, you can load any #GGUF model (#Llama, #Mistral, #CodeLlama, etc.) with zero extra setup.

It’s literally import #llm and go.

Using #codellama to generate C++ code that I've already written myself to see how it compares. I'm impressed by the simplicity of the code generated. I also don't want to beg it to fix the errors in the code. My opinion that #AI should be treated like #wikipedia hasn't changed. Trust but verify

📢New Feed "AI Coding 🤖"

"bsky.app/profile/did:plc:g5rkcuia...

Feed based on the following hashtags: #cline, #roocode, #cursor, #windsurf, #ai-engineering, #bolt, #loveable, #v0, #replit, #sagemaker, #codellama, #codewhisperer, #supermaven, #sourcegraph

Original post on fosstodon.org

#Ollama models:
➡️ #CodeLlama offers versatile development assistance with advanced pattern recognition and infilling capabilities.
➡️ Qwen 2.5 Coder transforms debugging by providing intelligent solutions to root causes.
➡️ WizardLM2 excels in rapid prototyping and multilingual […]

Meta’s free Code Llama AI programming tool closes the gap with GPT-4 OpenAI’s code generators have some new competition.

🤖 Meta AI dropped Code Llama 70B, a coding AI that can write better code than any model before. It even understands instructions in natural language! Coding just got a whole lot easier. 🤯
#AI #coding #CodeLlama #MetaAI
www.theverge.com/202...

Introducing Code Llama, an AI Tool for Coding | Meta We're releasing Code Llama, an AI model built on top of Llama 2 that's fine-tuned for generating and discussing code.

Meta released #CodeLlama, a large language model (#LLM) that can use text prompts to generate and discuss code. Code Llama is state-of-the-art for publicly available LLMs on coding tasks.
about.fb.com/news/2023/08...