CompassJudger-2 32B on Apple M3 Ultra (512 GB)

Can the Apple M3 Ultra (512 GB) run CompassJudger-2 32B?

Barely: it runs very slowly

Very slow. At Q8_0 it needs about 66.6 GB of 384 GB usable and makes one decision in about 13.1 s.

Best quant: Q8_0
Time per decision: 13.1 s
Decisions per second: 0.1/s
Input assumed: 1,024 tokens
Returns: Written feedback, then a score

Best app to run it on the Apple M3 Ultra (512 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX

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

Open CompassJudger-2 32B 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 CompassJudger-2 32B

Recommended for the Apple M3 Ultra (512 GB): Q8_0. 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 Q8_0 file (opens in a new tab)

huggingface.co/mradermacher/CompassJudger-2-32B-Instruct-GGUF (opens in a new tab)

ollama run hf.co/mradermacher/CompassJudger-2-32B-Instruct-GGUF:Q8_0

Original weights · Hugging Face

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

huggingface.co/opencompass/CompassJudger-2-32B-Instruct (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 86Excellent quality (S) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q8_0, from its editorial quality tier scaled by the quantization or precision. It’s the same on any hardware that runs this variant. Scale: S 85+ · A 70+ · B 55+ · C 40+ · D under 40.
Runs Very slow: How well it runs on the Apple M3 Ultra (512 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.	38.7 / 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.	1.3 / 15
ClarityHow easy the answer is to use in code: a probability or score earns the most, written feedback before a score the least.	9.0 / 15
Match score	74.0 / 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 (512 GB)

Quant	Memory needed	Available	Time per decision	Runs	Quality	Download
`Q4_K_M`	51.8 GB	384 GB	7.6 s	Very slow	81Great quality (A) · 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	55.1 GB	384 GB	8.8 s	Very slow	83Great quality (A) · 83 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q6_K`	58.7 GB	384 GB	10.1 s	Very slow	85Excellent quality (S) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	66.6 GB	384 GB	13.1 s	Very slow	86Excellent quality (S) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	97.4 GB	384 GB	26.1 s	Very slow	86Excellent quality (S) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.	Original weights (opens in a new tab)

Times are for one request with a 1,024-token input and about 150 tokens written. Batching servers process many requests at once.

Other hardware that runs it

85Excellent quality (S) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA B200 Q6_K · 1.0 s · Slow
85Excellent quality (S) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI350X Q6_K · 1.2 s · Slow
83Great quality (A) · 83 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI325X Q5_K_M · 1.3 s · Slow
81Great quality (A) · 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 141GB SXM Q4_K_M · 1.2 s · Slow
81Great quality (A) · 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. AMD Instinct MI300X Q4_K_M · 1.2 s · Slow
81Great quality (A) · 81 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA H200 NVL Q4_K_M · 1.2 s · Slow

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.