Qianfan-OCR 4B on Apple M2 Ultra (192 GB)

Can the Apple M2 Ultra (192 GB) run Qianfan-OCR 4B?

Yes, it runs great

Runs great. At F16 it needs about 22.8 GB of 144 GB usable, with an estimated speed of 41.9 tok/s.

Best quant: F16
Generation speed: 41.9 tok/s
Prompt processing: 3,558 tok/s
First token (1K prompt): 0.3 s
Largest safe context: 32K

Best app to run it on the Apple M2 Ultra (192 GB)

On macOS

LM Studio

Desktop app · runs on llama.cpp, MLX

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

Open Qianfan-OCR 4B 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 Qianfan-OCR 4B

Recommended for the Apple M2 Ultra (192 GB): F16. 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 F16 file (opens in a new tab)

huggingface.co/ggml-org/Qianfan-OCR-GGUF (opens in a new tab)

ollama run hf.co/ggml-org/Qianfan-OCR-GGUF:F16

Original weights · Hugging Face

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

huggingface.co/baidu/Qianfan-OCR (opens in a new tab)

No sign-in needed.

Official repo · GitHub

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

github.com/baidubce/Qianfan-VL (opens in a new tab)

The model card links Baidu's Qianfan-VL repo as Qianfan-OCR's official code.

git clone https://github.com/baidubce/Qianfan-VL

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

How it’s rated

Quality 50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at F16, 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 (192 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.	22.5 / 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.	3.7 / 15
Match score	66.2 / 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 (192 GB)

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	15.2 GB	144 GB	293 tok/s	Runs great	35Basic quality (D) · 35 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q3_K_M`	15.7 GB	144 GB	231 tok/s	Runs great	42Fair quality (C) · 42 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q4_K_M`	16.2 GB	144 GB	178 tok/s	Runs great	47Fair quality (C) · 47 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q5_K_M`	16.7 GB	144 GB	147 tok/s	Runs great	48Fair quality (C) · 48 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q6_K`	17.2 GB	144 GB	124 tok/s	Runs great	49Fair quality (C) · 49 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	GGUF files (opens in a new tab)
`Q8_0`	18.4 GB	144 GB	90.8 tok/s	Runs great	50Fair quality (C) · 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	22.8 GB	144 GB	41.9 tok/s	Runs great	50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab) 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

50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4070 Ti SUPER F16 · 40.8 tok/s · Runs great
50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4080 SUPER F16 · 44.7 tok/s · Runs great
50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5070 Ti F16 · 54.4 tok/s · Runs great
50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5080 F16 · 58.3 tok/s · Runs great
49Fair quality (C) · 49 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 GeForce RTX 4060 Q6_K · 48.9 tok/s · Runs great
50Fair quality (C) · 50 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. Intel Arc B580 Q8_0 · 45.9 tok/s · Runs great

Quick answers

How much memory does Qianfan-OCR 4B need on the M2 Ultra 192GB?

At F16 it needs about 22.8 GB. The M2 Ultra 192GB has 144 GB usable.

How fast is Qianfan-OCR 4B on the M2 Ultra 192GB?

About 41.9 tok/s at F16, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Qianfan-OCR 4B should I use on the M2 Ultra 192GB?

F16. 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.