Check my GPU

on

Can the Apple M2 Pro (32 GB) run Falcon-H1 Tiny 90M Instruct?

44 ptsUsable (C) · 44 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.8/15

Yes, it runs well

Runs well. At F16 it needs about 2.6 GB of 21.4 GB usable, with an estimated speed of 551 tok/s.

Best quant
F16
Generation speed
551 tok/s
Prompt processing
46,844 tok/s
First token (1K prompt)
0.3 s
Largest safe context
256K

Download Falcon-H1 Tiny 90M Instruct

Recommended for the Apple M2 Pro (32 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/tiiuae/Falcon-H1-Tiny-90M-Instruct-GGUF (opens in a new tab)

ollama run hf.co/tiiuae/Falcon-H1-Tiny-90M-Instruct-GGUF:BF16

Original weights · Hugging Face

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

huggingface.co/tiiuae/Falcon-H1-Tiny-90M-Instruct (opens in a new tab)

No sign-in needed.

Links checked 6 days ago against Hugging Face, Ollama and GitHub.

How the C grade is scored

Every result gets up to 100 points from four parts. The points decide the letter grade.

Model qualityEditorial quality tier of the model, scaled by the quantization or precision used.1.4 / 45
Memory fitFull marks when everything sits in GPU or unified memory; fewer when part spills to system RAM.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.2.8 / 15
Total44.1 / 10044 ptsUsable (C) · 44 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.8/15

Grade scale: S 85+ · A 72+ · B 58+ · C 42+ · D 25+ · F under 25 points.

Every quantization on the Apple M2 Pro (32 GB)

Quant Memory needed Available Speed Fit Grade · points Download
Q2_K 2.5 GB 21.4 GB 3,858 tok/s Runs well 43 ptsUsable (C) · 43 of 100 points: Model quality 0.9/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15 GGUF (opens in a new tab)
Q3_K_M 2.5 GB 21.4 GB 3,037 tok/s Runs well 43 ptsUsable (C) · 43 of 100 points: Model quality 1.1/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.6/15 GGUF (opens in a new tab)
Q4_K_M 2.5 GB 21.4 GB 2,349 tok/s Runs well 43 ptsUsable (C) · 43 of 100 points: Model quality 1.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15 GGUF (opens in a new tab)
Q5_K_M 2.5 GB 21.4 GB 1,941 tok/s Runs well 43 ptsUsable (C) · 43 of 100 points: Model quality 1.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15 GGUF (opens in a new tab)
Q6_K 2.5 GB 21.4 GB 1,630 tok/s Runs well 44 ptsUsable (C) · 44 of 100 points: Model quality 1.3/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15 GGUF (opens in a new tab)
Q8_0 2.5 GB 21.4 GB 1,196 tok/s Runs well 44 ptsUsable (C) · 44 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.7/15 GGUF (opens in a new tab)
F16 2.6 GB 21.4 GB 551 tok/s Runs well 44 ptsUsable (C) · 44 of 100 points: Model quality 1.4/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 2.8/15 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

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.