Aya Expanse 32B on NVIDIA RTX 4000 SFF Ada Generation

Can the NVIDIA RTX 4000 SFF Ada Generation run Aya Expanse 32B?

Yes, it runs well (tight fit)

Runs well (tight fit). At Q3_K_M it needs about 18.8 GB of 20.0 GB usable, with an estimated speed of 13.8 tok/s.

Best quant: Q3_K_M
Generation speed: 13.8 tok/s
Prompt processing: 428 tok/s
First token (1K prompt): 2.4 s
Largest safe context: 15K

Best app to run it on the NVIDIA RTX 4000 SFF Ada Generation

On Windows

Ollama

Command line + local API · runs on llama.cpp (ggml), MLX (Apple silicon, preview)

One command with Ollama’s own build of this model.
Serves an OpenAI-compatible API that other apps and coding tools can use.

ollama run aya-expanse:32b-q3_K_M

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

On Linux

Ollama

Command line + local API · runs on llama.cpp (ggml), MLX (Apple silicon, preview)

One command with Ollama’s own build of this model.
Serves an OpenAI-compatible API that other apps and coding tools can use.

ollama run aya-expanse:32b-q3_K_M

LM Studio Point-and-click: search, download and chat in one window.
llama.cpp Usually the first to support new model architectures.

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 Aya Expanse 32B

Recommended for the NVIDIA RTX 4000 SFF Ada Generation: Q3_K_M. 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 aya-expanse:32b-q3_K_M

Quantized files · GGUF

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

Download the Q3_K_M (Q3_K_L) file (opens in a new tab)

huggingface.co/lmstudio-community/aya-expanse-32b-GGUF (opens in a new tab)

ollama run hf.co/lmstudio-community/aya-expanse-32b-GGUF:Q3_K_L

Original weights · Hugging Face

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

huggingface.co/CohereLabs/aya-expanse-32b (opens in a new tab)

Free Hugging Face account needed: accept the license on the model page and the download starts right away.

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 Q3_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.: How capable the model itself is at Q3_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 welltight fit: How well it runs on the NVIDIA RTX 4000 SFF Ada Generation: 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.	22.7 / 45
Memory fitFull marks when everything sits in GPU memory; fewer when part spills to system RAM.	24.0 / 25
SpeedGeneration speed against a 40 tokens/s target for LLMs, or render time against a target for image and video.	8.8 / 15
Memory useRewards using the memory you have, so a bigger, better variant beats a tiny one that leaves most memory idle.	15.0 / 15
Match score	70.5 / 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 NVIDIA RTX 4000 SFF Ada Generation

Quant	Memory needed	Available	Speed	Runs	Quality	Download
`Q2_K`	15.9 GB	20.0 GB	17.5 tok/s	Runs well	42Fair quality (C) · 42 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.
`Q3_K_M`	18.8 GB	20.0 GB	13.8 tok/s	Runs welltight fit	50Fair quality (C) · 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (Q3_K_L) (opens in a new tab)
`Q4_K_M`	22.6 GB	20.0 GB	7.4 tok/s	Slowoffloads to RAM	57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q5_K_M`	25.9 GB	20.0 GB	4.8 tok/s	Slowoffloads to RAM	58Good quality (B) · 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 56/100 (quality, fit, speed and memory use) sets the ranking.
`Q6_K`	29.4 GB	20.0 GB	3.5 tok/s	Slowheavy RAM offload	59Good quality (B) · 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`Q8_0`	37.2 GB	20.0 GB	2.1 tok/s	Very slowheavy RAM offload	60Good quality (B) · 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 49/100 (quality, fit, speed and memory use) sets the ranking.	GGUF (opens in a new tab)
`F16`	67.5 GB	20.0 GB	—	Won't run	60Good quality (B) · 60 of 100: how capable the model is at F16, whatever it runs on. Match score 8/100 (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

58Good quality (B) · 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 5090 Q5_K_M · 56.3 tok/s · Runs great
57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 4090 Q4_K_M · 38.3 tok/s · Runs well (tight fit)
57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA GeForce RTX 3090 Q4_K_M · 35.6 tok/s · Runs well (tight fit)
57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking. AMD Radeon RX 7900 XTX Q4_K_M · 31.4 tok/s · Runs well (tight fit)
59Good quality (B) · 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. NVIDIA RTX PRO 6000 Blackwell Q6_K · 47.3 tok/s · Runs great
57Good quality (B) · 57 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. Apple M4 Max (64 GB) Q4_K_M · 17.9 tok/s · Runs well

Quick answers

How much memory does Aya Expanse 32B need on the NVIDIA RTX 4000 SFF Ada Generation 20GB?

At Q3_K_M it needs about 18.8 GB. The NVIDIA RTX 4000 SFF Ada Generation 20GB has 20.0 GB usable.

How fast is Aya Expanse 32B on the NVIDIA RTX 4000 SFF Ada Generation 20GB?

About 13.8 tok/s at Q3_K_M, an estimate from the hardware’s memory bandwidth and the size of the model.

Which quantization of Aya Expanse 32B should I use on the NVIDIA RTX 4000 SFF Ada Generation 20GB?

Q3_K_M. 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.