NuExtract 2.0 2B on Apple M3 Ultra (256 GB)

Can the Apple M3 Ultra (256 GB) run NuExtract 2.0 2B?

Yes, it runs great

Runs great. At Q4_K_M it needs about 17.4 GB of 192 GB usable and makes one decision in about 97 ms.

Best quant: Q4_K_M
Time per decision: 97 ms
Decisions per second: 10/s
Input assumed: 256 tokens
Returns: Labeled spans or fields

Best app to run it on the Apple M3 Ultra (256 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX

Point-and-click: search, download and chat in one window.

Open NuExtract 2.0 2B in LM Studio

llama.cpp Usually the first to support new model architectures.
Ollama Runs the Hugging Face GGUF with ollama run hf.co/…

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 NuExtract 2.0 2B

Recommended for the Apple M3 Ultra (256 GB): Q4_K_M. Every other quantization has its own download link in the table below.

Quantized files · GGUF

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

Download the Q4_K_M file (opens in a new tab)

huggingface.co/numind/NuExtract-2.0-2B-GGUF (opens in a new tab)

ollama run hf.co/numind/NuExtract-2.0-2B-GGUF:Q4_K_M

Original weights · Hugging Face

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

huggingface.co/numind/NuExtract-2.0-2B (opens in a new tab)

No sign-in needed.

Links checked 3 days ago against Hugging Face, Ollama and GitHub.

How it’s rated

Quality 66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q4_K_M, 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 M3 Ultra (256 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.

PrecisionEditorial precision tier from the model’s published benchmarks, scaled by the quantization or precision used.	29.9 / 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
SpeedTime for one decision against a 100 ms target: reading the input plus any tokens the model writes.	15.0 / 15
ClarityHow easy the answer is to use in code: a probability or score earns the most, written feedback before a score the least.	13.0 / 15
Match score	82.9 / 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 M3 Ultra (256 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q4_K_M`	17.4 GB	192 GB	97 ms	Runs great	66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	17.6 GB	192 GB	110 ms	Runs great	67Good quality (B) · 67 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	17.8 GB	192 GB	124 ms	Runs great	69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	18.4 GB	192 GB	156 ms	Runs well	70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	20.4 GB	192 GB	297 ms	Runs well	70Great quality (A) · 70 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab) Original weights (opens in a new tab)

Times are for one request with a 256-token input and about 24 tokens written. Batching servers process many requests at once.

Other hardware that runs it

70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 Q8_0 · 98 ms · Runs great
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q8_0 · 59 ms · Runs great
70Great quality (A) · 70 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q8_0 · 58 ms · Runs great
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 7900 XTX Q6_K · 87 ms · Runs great
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3090 Q6_K · 93 ms · Runs great
69Good quality (B) · 69 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5070 Ti Q6_K · 91 ms · Runs great

Quick answers

How much memory does NuExtract 2.0 2B need on the M3 Ultra 256GB?

At Q4_K_M it needs about 17.4 GB. The M3 Ultra 256GB has 192 GB usable.

How fast is NuExtract 2.0 2B on the M3 Ultra 256GB?

It makes one decision in about 97 ms at Q4_K_M, an estimate from the hardware and the size of the model.

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.