Llama Meta Released 2024-07 chatinstructiontools

Llama 3.1

Meta's flagship open-weights LLM. Strong all-rounder with native tool use, 128K context, and broad ecosystem support.

Best for general-purpose chat, agentic workflows, code
Sizes 8B · 70B · 405B
Context 128K
License Llama 3.1 Community License
Min VRAM (default size, Q4) 5 GB
Rec VRAM 8 GB

What is Llama 3.1?

Llama 3.1 is Meta's open-weights language model. Released in 2024-07, it ships in 3 sizes (8B, 70B, 405B) and is licensed Llama 3.1 Community License. The most popular use case is general-purpose chat, agentic workflows, code.

VRAM and hardware

The smallest 8B size needs at least 5 GB VRAM; the largest 405B needs around 5 GB.

Size Min VRAM (Q4_K_M) Recommended VRAM Tokens/sec on 3090
8B 5 GB 8 GB ~25 tok/s
70B 42 GB 63 GB ~14 tok/s
405B 240 GB 360 GB ~10 tok/s

How to run Llama 3.1 locally

Option 1: Ollama (simplest)

ollama pull llama-3-1
ollama run llama-3-1

Option 2: Mullama (production)

mullama pull llama-3-1
mullama run llama-3-1

Option 3: llama.cpp (CLI)

# Download a GGUF from Hugging Face (search "llama-3-1 gguf")
./llama-cli -m llama-3-1.Q4_K_M.gguf -p "Hello, AI!"

Option 4: Python with Mullama or llama-cpp-python

from mullama import Model, Context
model = Model.load("llama-3-1.Q4_K_M.gguf", n_gpu_layers=99)
ctx = Context(model, n_ctx=4096)
print(ctx.generate("Hello, AI!", 256))

What you can build with Llama 3.1

  • chat
  • code
  • RAG
  • agentic
  • summarization

Hardware it fits on

  • Apple Silicon (8GB): 8B in MLX
  • Apple Silicon (16GB): 70B in MLX
  • Apple Silicon (32GB): 405B in MLX or GGUF
  • Apple Silicon (64GB+): 405B in MLX
  • RTX 3090 (24GB): 405B at Q4_K_M
  • RTX 4090 (24GB): same as 3090, ~30% faster
  • RTX 5090 (32GB): 405B at Q8_0
  • 2× RTX 6000 Ada (96GB total): 405B at Q4_K_M

Related models in the Llama family

  • Llama 3.2 — 1B, 3B · low-end hardware, mobile, edge deployment
  • Llama 3.2 Vision — 11B, 90B · vision + chat on desktop GPUs
  • Llama 3.3 — 70B · best-in-class 70B for desktops
  • Llama 4 — Scout 17B (109B MoE), Maverick 17B (400B MoE) · vision + long context (10M tokens)

Other models you might consider

  • Qwen 2.5 (Alibaba) — 0.5B, strongest 7B-32B on consumer hardware
  • Qwen 2.5-Coder (Alibaba) — 0.5B, best open-weight code model in 2024-2025
  • Qwen 3 (Alibaba) — 0.6B, best all-round local LLM in 2025-2026
  • Qwen 3-Coder (Alibaba) — 30B, long-context code agent workflows
  • DeepSeek R1 (DeepSeek) — 1.5B, reasoning, math, step-by-step problem solving
  • DeepSeek V3 (DeepSeek) — 671B (37B active MoE), frontier-class open-weights LLM (needs data-center GPUs)