bitnet-b1.58-2b-4t
A Core AI bundle of microsoft/bitnet-b1.58-2B-4T for the Aether SDK (iOS and macOS 27+).
- Source:
microsoft/bitnet-b1.58-2B-4Tat revision04c3b9ad9361b824064a1f25ea60a8be9599b127, licence MIT. The licence is included asLICENSE. - Changes from the source: converted from PyTorch to Core AI (
.aimodel) by Aether forge (recipebitnet-b1.58-2b-4t@1). Weights are int8-linear-perchannel (8-bit weights). The tokenizer files are the source's own.
Variants
| Variant | Platform | Arch | Compute | Compiled | Assets | Download |
|---|---|---|---|---|---|---|
macos-any-gpu |
macos | any | gpu | no (specialized on first load) | bitnet_b1_58_2b_4t.aimodel 2.42 GB |
2.42 GB |
ios-any-gpu |
ios | any | gpu | no (specialized on first load) | bitnet_b1_58_2b_4t.aimodel 2.42 GB |
2.42 GB |
ios-h18p-gpu |
ios | h18p | gpu | yes | bitnet_b1_58_2b_4t.h18p.aimodelc 2.42 GB |
2.42 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 | 24A446 | target | 5417ccda |
|
ios-any-gpu |
T2 | pass | 19/19 strict; profile quantized-8bit; fixture 71736ee9acc69fd0 |
iPhone18,2 | 24A446 | target | c7014699 |
ios-h18p-gpu |
T0 | pass | iPhone18,2 | 24A446 | target | 5f91bbca |
|
ios-h18p-gpu |
T2 | pass | 19/19 strict; profile quantized-8bit; fixture 71736ee9acc69fd0 |
iPhone18,2 | 24A446 | target | 6435bf43 |
macos-any-gpu |
T0 | pass | Mac17,6 | 26A434 | target | 16c37e56 |
|
macos-any-gpu |
T1 | pass | Mac17,6 | 26A434 | target | 235ebbf7 |
|
macos-any-gpu |
T2 | pass | 19/19 strict; 19/19 strict; profile quantized-8bit; fixture 71736ee9acc69fd0 |
Mac17,6 | 26A434 | target | a359dd63 |
| unquantized reference (not published) | T2 | pass | 19/19 strict; 19/19 strict; profile strict; fixture 71736ee9acc69fd0 |
Mac17,6 | 26A434 | target | 3b8f311c |
The quantized profile also requires: T2 strict on the unquantized reference export (met by the reference row).
Use
aether run bitnet-b1.58-2b-4t --prompt "Hello"
import Aether
let aether = try Aether()
let chat = try await aether.chat("bitnet-b1.58-2b-4t")
let reply = try await chat.respond(to: "Hello")
print(reply.text)
Model tree for aether-models/bitnet-b1.58-2b-4t
Base model
microsoft/bitnet-b1.58-2B-4T