OpenModelsHQ

aya-expanse-32b

Cohere LabsLLMDenseCC BY-NC 4.0

aya-expanse-32b is an open-weight language model from Cohere Labs with 32.3B parameters. At 4-bit it needs about 23 GB of memory, so the smallest current device that fits it is the Radeon AI PRO R9700.

Overview

Estimated memory at 4-bit
23GB

Includes room for an 8k context and runtime overhead. More context needs more memory.

Total parameters
32.3B
Dense
Context window
Unknown
Not in the model config
Cheapest API, blended / 1M
No hosted API
Weights only

Key information

Lab
Cohere Labs
Released
23 Oct 2024
Parameters
32.3B
Type
Language
Context
Unknown
HF downloads, 30 days
3,376
View all

Will it run on your hardware?

How we estimate
Memory needed vs devices
aya-expanse-32b needs about 23 GB at 4-bit, 40.7 GB at 8-bit and 76.7 GB at 16-bit. At 4-bit it fits GeForce RTX 5090, DGX Spark, Mac Studio M5 Ultra (256GB). Dashed lines are usable memory on each device; a model fits when its estimate stays within 90% of it. Log scale. 8k context.
Memory needed vs devices: data
Estimate or deviceMemory
Q423 GB
Q840.7 GB
BF1676.7 GB
RTX 5060 Ti 16 GB16 GB usable
RTX 5090 32 GB32 GB usable
DGX Spark 128 GB108.8 GB usable
Mac Studio M5 Ultra 256 GB192 GB usable

Single-device memory estimate. Pick a device to update the On your device column.

36 GB usable memory

FormatMemory neededOn your deviceExample hardware
4-bit (Q4)≈ 23 GBFits 64%Radeon AI PRO R970032 GB memory
8-bit (Q8)≈ 40.7 GBToo big 113%L40S48 GB memory
16-bit (BF16)≈ 76.7 GBToo big 213%RTX PRO 6000 Blackwell96 GB memory

Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.

Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.

Where to run it

On a Mac

Start with the MLX files, built for Apple Silicon.

mlx-community/aya-expanse-32b-8bitCheck your Mac

With GGUF

Choose a file for llama.cpp or a compatible app.

bartowski/aya-expanse-32b-GGUFFind compatible hardware

Through an API

No hosted API found.

MLX on Apple SiliconView repository
mlx_lm.generate --model mlx-community/aya-expanse-32b-8bit \
  --prompt "Explain how a transformer works."
GGUF with llama.cppView repository
llama-cli -hf bartowski/aya-expanse-32b-GGUF \
  -c 8192 -p "Explain how a transformer works."

Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.

More about this model

Frequently asked questions

How much memory does aya-expanse-32b need? At an 8k context, about 23 GB at 4-bit, 40.7 GB at 8-bit and 76.7 GB at 16-bit (BF16). A longer context needs more.

Can I run aya-expanse-32b on a Mac? The smallest current Mac that fits the 4-bit estimate of 23 GB is the Mac mini M5 Pro (48GB). An MLX build is available: mlx-community/aya-expanse-32b-8bit.

Does the full context fit? We do not have a context length for this model. Our memory estimates use 8k, so long prompts need more memory than shown.

What does it cost through an API? We did not find a hosted API for this model.

License in plain English

CC BY-NC 4.0. This license allows sharing and adapting with credit, but not for commercial purposes. This summary is not legal advice.

Read CC BY-NC 4.0

OMHQ RigBench

Not yet tested. We show no speed or quality score without a measurement.

Revision history and sources

Last updated . Sources:Hugging Face model card, MLX repository, GGUF repository.