granite-4.0-1b
A Core AI bundle of ibm-granite/granite-4.0-1b for the Aether SDK (iOS and macOS 27+).
- Source:
ibm-granite/granite-4.0-1bat revision6a7381ba1f54d684ff508d991aeb7dc580157103, licence Apache-2.0. The licence is included asLICENSE.ibm-granite/granite-4.0-1bdeclares apache-2.0 in its model card but ships no licence file;LICENSEis the canonical text from apache.org. - Changes from the source: converted from PyTorch to Core AI (
.aimodel) by Aether forge (recipegranite-4.0-1b@2). Weights are int8-linear-perblock32 (8-bit weights). The tokenizer files are the source's own.
Variants
| Variant | Platform | Arch | Compute | Compiled | Download |
|---|---|---|---|---|---|
macos-any-gpu |
macos | any | gpu | no (specialized on first load) | 1.74 GB |
ios-any-gpu |
ios | any | gpu | no (specialized on first load) | 1.74 GB |
ios-h18p-gpu |
ios | h18p | gpu | yes | 1.74 GB |
Verification
Every row is a record in verification/ about exactly these bytes (matched by bundle digest). Reference rows
are strict T2 passes of the unquantized export on the same fixture, in verification/reference/.
| Variant | Tier | Result | Detail | Device | OS build | Compute | Record |
|---|---|---|---|---|---|---|---|
ios-any-gpu |
T0 | pass | iPhone18,2 | 24A437 | target | 70a17ed2 |
|
ios-any-gpu |
T2 | pass | 19/19 strict; profile quantized-8bit; fixture aa4c544dc96a983e |
iPhone18,2 | 24A437 | target | 29cba282 |
ios-h18p-gpu |
T0 | pass | iPhone18,2 | 24A437 | target | 8bba98be |
|
ios-h18p-gpu |
T2 | pass | 19/19 strict; profile quantized-8bit; fixture aa4c544dc96a983e |
iPhone18,2 | 24A437 | target | b28eeafc |
macos-any-gpu |
T0 | pass | Mac17,6 | 26A428 | target | e9b192a2 |
|
macos-any-gpu |
T1 | pass | Mac17,6 | 26A428 | target | e510b868 |
|
macos-any-gpu |
T2 | pass | 19/19 strict; 19/19 strict; profile quantized-8bit; fixture aa4c544dc96a983e |
Mac17,6 | 26A428 | target | 14c59d9c |
| unquantized reference (not published) | T2 | pass | 19/19 strict; 19/19 strict; profile strict; fixture aa4c544dc96a983e |
Mac17,6 | 26A428 | target | 4c32f671 |
The quantized profile also requires: T2 strict on the unquantized reference export (met by the reference row).
Use
aether run granite-4.0-1b --prompt "Hello"
import Aether
let aether = try Aether()
let chat = try await aether.chat("granite-4.0-1b")
let reply = try await chat.respond(to: "Hello")
print(reply.text)
Model tree for aether-models/granite-4.0-1b
Base model
ibm-granite/granite-4.0-1b-base Finetuned
ibm-granite/granite-4.0-1b