Code Llama 13B Instruct on Apple M2 Ultra (128 GB)

Can the Apple M2 Ultra (128 GB) run Code Llama 13B Instruct?

Yes, it runs great

Runs great. At Q6_K it needs about 25.7 GB of 96.0 GB usable, with an estimated speed of 45.1 tok/s.

Best quant: Q6_K
Generation speed: 45.1 tok/s
Prompt processing: 1,295 tok/s
First token (1K prompt): 0.8 s
Largest safe context: 16K

Best app to run it on the Apple M2 Ultra (128 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX · use the MLX build

Runs the MLX build, typically 10–20% faster than GGUF for a model with 13B active parameters on Apple silicon.
Point-and-click: search, download and chat in one window.

Open Code Llama 13B Instruct in LM Studio

Ollama One command with Ollama’s own build of this model.
llama.cpp Usually the first to support new model architectures.

Want an agent? Code Llama 13B Instruct supports tool use, so you can pair a llama.cpp or MLX LM server with one of these. Apps that run on this computer come first.

Coding agents

Personal assistants

How we choose: the files published for this model, your operating system and GPU, whether it fits fully in memory, and published speed tests (MLX vs llama.cpp on Apple silicon, vLLM vs llama.cpp vs Ollama on NVIDIA). Details.

Download Code Llama 13B Instruct

Recommended for the Apple M2 Ultra (128 GB): Q6_K. Every other quantization has its own download link in the table below.

Quickest · Ollama

One command downloads a ready-to-run version and starts a chat.

Get Ollama at ollama.com/download (opens in a new tab) · model page (opens in a new tab)

ollama run codellama:13b-instruct-q6_K

Quantized files · GGUF

Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.

Download the Q6_K file (opens in a new tab)

huggingface.co/TheBloke/CodeLlama-13B-Instruct-GGUF (opens in a new tab)

ollama run hf.co/TheBloke/CodeLlama-13B-Instruct-GGUF:Q6_K

Original weights · Hugging Face

Full-precision files from Meta, for fine-tuning, vLLM, Transformers.

huggingface.co/meta-llama/CodeLlama-13b-Instruct-hf (opens in a new tab)

Free Hugging Face account needed: request access on the model page, and Meta approves it (usually within minutes to a few hours).

Official repo · GitHub

Reference code, inference scripts and docs from Meta. The weights themselves download from Hugging Face.

github.com/meta-llama/codellama (opens in a new tab)

Meta’s official Code Llama repo.

git clone https://github.com/meta-llama/codellama

Links checked 1 day ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 29Basic quality (D) · 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q6_K, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: A+ 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Runs great: How well it runs on the Apple M2 Ultra (128 GB): speed, capped by how well it fits in memory. Runs great: 30+ tokens/s (or near the image, video or decision speed target) · Runs well: 10+ tokens/s · Slow: 3+ tokens/s · Very slow: under 3 · Won’t run: doesn’t fit even with offload. A tight fit or RAM offload caps it at Runs well; heavy offload caps it at Slow.

Match score

To order lists, each result also gets a match score out of 100 from four parts. It favors the most capable model that still runs well.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.	13.4 / 45
Memory fitFull marks when the model fits in the share of unified memory the GPU can use; fewer when it goes past that limit.	25.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	15.0 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	6.2 / 15
Match score	59.5 / 100

The match score orders results; it is not shown as a grade. Quality and Runs are the two ratings in the tables.

Every quantization on the Apple M2 Ultra (128 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	20.2 GB	96.0 GB	107 tok/s	Runs great	21Basic quality (D) · 21 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 54/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q3_K_M`	21.4 GB	96.0 GB	84.0 tok/s	Runs great	25Basic quality (D) · 25 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q4_K_M`	22.9 GB	96.0 GB	64.9 tok/s	Runs great	28Basic quality (D) · 28 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	24.2 GB	96.0 GB	53.7 tok/s	Runs great	29Basic quality (D) · 29 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	25.7 GB	96.0 GB	45.1 tok/s	Runs great	29Basic quality (D) · 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	28.8 GB	96.0 GB	33.1 tok/s	Runs great	30Basic quality (D) · 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	41.0 GB	96.0 GB	15.2 tok/s	Runs well	30Basic quality (D) · 30 of 100: how capable the model is at F16, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Memory figures assume 8K of context. Longer conversations or documents grow the KV cache.

Other hardware that runs it

30Basic quality (D) · 30 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q8_0 · 86.0 tok/s · Runs great
29Basic quality (D) · 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 7900 XTX Q6_K · 54.1 tok/s · Runs great
29Basic quality (D) · 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3090 Q6_K · 61.2 tok/s · Runs great
29Basic quality (D) · 29 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 Q6_K · 65.9 tok/s · Runs great
25Basic quality (D) · 25 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 9070 XT Q3_K_M · 67.2 tok/s · Runs well (tight fit)
25Basic quality (D) · 25 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4070 Ti SUPER Q3_K_M · 81.9 tok/s · Runs well (tight fit)

Quick answers

How much memory does Code Llama 13B Instruct need on the M2 Ultra 128GB?

At Q6_K it needs about 25.7 GB. The M2 Ultra 128GB has 96.0 GB usable.

How fast is Code Llama 13B Instruct on the M2 Ultra 128GB?

About 45.1 tok/s at Q6_K, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Code Llama 13B Instruct should I use on the M2 Ultra 128GB?

Q6_K. Of the versions that fit, it is the one that runs best; when two run equally well, the check prefers the higher-rated one.

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.