Models in Apple’s Core AI catalogue. Apple publishes export recipes in apple/coreai-models that convert each one to run on-device.
yolos-tiny
Builds
—
Memory needed (est.)
—
Quantisation
—
Devices
Runs on: Mac, iPhone
No memory estimate: size not published
8
16
24
32
64
128
apache-2.0Added 8 Jun 2026
clap-htsat-unfused
Builds
—
Memory needed (est.)
—
Quantisation
—
Devices
Runs on: Mac, iPhone
No memory estimate: size not published
8
16
24
32
64
128
apache-2.0Added 8 Jun 2026
Muse-Glimmer-30B-assistant
Context
128K tokens
Memory needed (est.)
—
Quantisation
None
Devices
Runs on: Mac
No memory estimate: size not published
8
16
24
32
64
128
apache-2.0Added 31 Aug 2026
Phi-3-mini-4k-instruct
Context
4K tokens
Memory needed (est.)
5 GB
Quantisation
4-bit
Devices
Runs on: Mac
Fits in unified memory
8
16
24
32
64
128
mitAdded 20 Aug 2026
Phi-3.5-mini-instruct
Context
128K tokens
Memory needed (est.)
5 GB
Quantisation
4-bit
Devices
Runs on: Mac
Fits in unified memory
8
16
24
32
64
128
mitAdded 20 Aug 2026
Mixtral-8x7B-Instruct-v0.1
Context
32K tokens
Memory needed (est.)
32 GB
Quantisation
4-bit
Devices
Runs on: Mac
Fits in unified memory
8
16
24
32
64
128
apache-2.0Added 8 Jun 2026
clip-vit-base-patch32
Builds
—
Memory needed (est.)
—
Quantisation
—
Devices
Runs on: Mac, iPhone
No memory estimate: size not published
8
16
24
32
64
128
Licence not statedAdded 8 Jun 2026
Qwen2.5-1.5B-Instruct
Context
32K tokens
Memory needed (est.)
3 GB
Quantisation
4-bit, 4-bit weights, 8-bit KV cache, 4-bit weight palettized group8
Devices
Runs on: Mac, iPhone
Fits in unified memory
8
16
24
32
64
128
apache-2.0Added 8 Jun 2026
Wan2.1-T2V-1.3B-Diffusers
Builds
—
Memory needed (est.)
—
Quantisation
None
Devices
Devices not stated
No memory estimate: size not published
8
16
24
32
64
128
apache-2.0Added 25 Aug 2026
Showing 57 of 57 · Previous page