Llama 3.3 70B
Meta's big open model, close to cloud models from a year or two ago. It needs about 45 GB, so two 24 GB cards or a Mac with 64 GB or more.
ollama run llama3.3:70bNeeds about 45 GB of graphics memory, or unified memory on a Mac. First time? Install Ollama, then paste that into a terminal.
Download43 GB
Memory to runAbout 45 GB
Read per token43 GB
LicenseLlama 3.3 Community License
Context128K tokens
ReleasedDecember 2024
Good for
- General knowledge and writing quality
- Following detailed instructions
- Multilingual chat
Not the best pick for
- Single consumer cards: it won't fit
- Speed on Macs, where it runs at a few tokens per second
25 GPUs and Macs
What runs Llama 3.3 70B
Estimated speed on each card at the smallest memory size that fits it, from memory bandwidth. As a dense model it reads all its weights for every token, so memory bandwidth sets the speed.
Runs great 2 cards
Usable 9 cards
Slow 14 cards
- RTX A600048 GB · ~10 tok/s
- A4048 GB · ~9 tok/s
- Quadro RTX 800048 GB · ~9 tok/s
- Apple M4 Max64 GB · ~7 tok/s
- Apple M5 Max64 GB · ~6 tok/s
- Apple M1 Max64 GB · ~5 tok/s
- Apple M2 Max64 GB · ~5 tok/s
- Apple M3 Max64 GB · ~5 tok/s
- Apple M5 Pro64 GB · ~4 tok/s
- Apple M4 Pro64 GB · ~3 tok/s
- NVIDIA DGX Spark128 GB · ~3 tok/s
- Ryzen AI Max+ 395 (Radeon 8060S)64 GB · ~3 tok/s
- Ryzen AI Max 390 (Radeon 8050S)64 GB · ~3 tok/s
- Jetson AGX Orin64 GB · ~3 tok/s
Cheapest build from our list that runs it well: Workstation + 2× RTX 3090, from $3,500 used, at about 12 tok/s. See the builds.
Learn
Read more about Llama 3.3 70B
- Buying a used RTX 3090 for AI: what to check and what to payWhy the RTX 3090 is still the default card for local AI, what a fair used price is, how to spot a bad one, and what to pair it with.
- Context length and the KV cache: why long chats use more VRAMWhy a model that fits can still run out of memory on a long document, how to work out the KV cache size, and three ways to shrink it.
- CPU offloading explained: running a model bigger than your VRAMWhat happens when a model doesn't fit in GPU memory, how partial offload works in Ollama and llama.cpp, how slow it gets, and when it's worth it.
- Does PCIe bandwidth matter for LLMs? x1 risers, x4 slots and x16When a narrow PCIe link slows a local model down and when it doesn't, from loading time to multi-GPU and CPU offload, and how to check your link.
- Free up disk space in Ollama: how to update and delete modelsSee how much space your Ollama models use, delete the ones you don't need, update the rest, and clean up after LM Studio and Hugging Face too.
- How much VRAM do you need to run an LLM locally?A simple rule for how much graphics memory a model needs, what fits in 8, 12, 16, 24 and 48 GB, and why context length adds more.
- Llama 3.3 70B hardware guide: what you need to run it locallyLlama 3.3 70B needs about 45 GB at 4-bit. The setups that run it, two 3090s, a 48 GB card or a big Mac, with estimated speeds and used prices.
- Local AI for small business: a private assistant for your teamWhat a small business can do with AI running on its own hardware, what it costs, which build to buy, and how to share one machine with the whole team.
- Open model licenses explained: Apache 2.0, Llama and Gemma termsWhat the licenses on popular open models mean in practice for personal, business and product use. Apache 2.0, the Llama community license and Gemma terms.
- Power supplies for multi-GPU AI rigs: how many watts you needHow to size a power supply for two or more GPUs, the cables each card needs, wall circuit limits, power capping, and what our builds draw under load.
- Qwen3 32B vs Llama 3.3 70B: is the bigger model worth it?Qwen3 32B fits one 24 GB card. Llama 3.3 70B needs about 45 GB. What you gain and lose with each, with estimated speeds and hardware costs.
- RTX 4090 vs two RTX 3090s for local AI: which should you buy?One 24 GB RTX 4090 or two used 3090s with 48 GB? Speeds, what each can run, power, cost and which is right for chat, 70B models and image generation.
- Running LLMs on a Mac: which M-series chip and how much memoryWhy Macs with Apple silicon are good at local AI, how much unified memory you need for each model size, and how they compare to a PC with a GPU.
- Speculative decoding: make local LLMs faster with a draft modelHow speculative decoding uses a small draft model to speed up a big one with no quality loss, when it helps, and how to turn it on in llama.cpp.
- The best local LLM for 32 GB, 48 GB and more (multi-GPU included)What to run with 32 GB, 48 GB or 64 GB+ of GPU memory, from an RTX 5090 to two 3090s to a server full of P100s. Picks, estimated speeds and our rack data.
- The best local LLM for a Mac with 64 GB, 96 GB or 128 GB+What to run on a Mac Studio or MacBook Pro with 64 GB or more of unified memory, from Qwen3 30B-A3B to Llama 3.3 70B and gpt-oss 120B, with estimated speeds.
- The best local LLM for writing, by how much memory you haveWhich open models write best for drafts, editing, emails and fiction on 8 GB to 48 GB+, plus the settings and prompts that make local writing sound less generic.
- The best local model for translationWhich open models translate best on your own hardware, by memory size, plus prompts that get natural results and where local translation still falls short.
- The RTX 5090 for local AI: what 32 GB and 1.8 TB/s get youHow fast the RTX 5090 runs local LLMs, what fits in 32 GB, what doesn't, the power you need, and when a used 3090 or two is the smarter buy.
- Two GPUs for local LLMs: layer split vs tensor parallel explainedWhat a second GPU gets you, why two cards usually run at one card's speed, when tensor parallel is faster, and what we measured on two-card rigs.
- Uncensored local models: what abliterated means, and the tradeoffsWhy local models refuse some requests, what abliterated and uncensored fine-tunes change, what it costs in quality, and how to cut needless refusals without one.
- Used workstations for local AI: Dell T7810, HP Z840 and similarWhy old Xeon workstations like the Dell T7810 and HP Z840 make cheap multi-GPU AI boxes, what to check before buying, and how to fit GPUs in them.
- Where Ollama stores models, and how to change the locationThe default Ollama model folder on Windows, Mac and Linux, how to move it to another drive with OLLAMA_MODELS, and how to bring your existing models along.
Compare
Other models
- Qwen3 4BAbout 4 GB · Quick everyday chat and summaries
- Qwen3 8BAbout 7 GB · A solid all-rounder for chat and writing
- Gemma 3 12BAbout 10 GB · Reads images as well as text
- Qwen3 14BAbout 11 GB · Noticeably smarter at reasoning and code
- gpt-oss 20BAbout 16 GB · OpenAI's open model. Fast, strong reasoning
- Gemma 3 27BAbout 19 GB · Google's best open model for its size
- Qwen3 30B-A3BAbout 21 GB · 30B smarts at small-model speed
- Qwen3 32BAbout 22 GB · Strong coding and long answers
- gpt-oss 120BAbout 67 GB · OpenAI's largest open model
- Qwen3 235B-A22BAbout 144 GB · Alibaba's flagship open model, for multi-GPU rigs
- DeepSeek R1 671BAbout 406 GB · The full DeepSeek R1, for the biggest rigs