Qwen3 1.7B

A Core AI bundle of Qwen/Qwen3-1.7B for the Aether SDK (iOS and macOS 27+).

  • Source: Qwen/Qwen3-1.7B at revision 70d244cc86ccca08cf5af4e1e306ecf908b1ad5e, licence Apache-2.0. The licence is included as LICENSE.
  • Changes from the source: converted from PyTorch to Core AI (.aimodel) by Aether forge (recipe qwen3-1.7b@2). Weights are int8-linear-perchannel (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 T2 pass 18/19 strict; profile quantized-8bit; budgeted: haiku-sea; fixture 13a4fa523908f7c0 iPhone18,2 24A437 target 7a11d3aa
ios-h18p-gpu T2 pass 18/19 strict; profile quantized-8bit; budgeted: haiku-sea; fixture 13a4fa523908f7c0 iPhone18,2 24A437 target 8b9f6cab
macos-any-gpu T2 pass 18/19 strict; profile quantized-8bit; budgeted: haiku-sea; fixture 13a4fa523908f7c0 Mac17,6 26A428 cpuOnly 1a8a270a
macos-any-gpu T3 pass gsm8k-test-100; 78.0% vs reference 73.0%; 100 items Mac17,6 26A428 gpu 0e805221
macos-any-gpu T3 pass copy-fidelity-v1; 100.0% vs reference 100.0%; 50 items Mac17,6 26A428 cpuOnly 2ddd6447
unquantized reference (not published) T2 pass 19/19 strict; profile strict; fixture 13a4fa523908f7c0 Mac17,6 26A428 gpu 8bd54932
unquantized reference (not published) T2 pass 19/19 strict; profile strict; fixture 13a4fa523908f7c0 iPhone18,2 24A437 target ef59f11b

The quantized profile also requires: T2 strict on the unquantized reference export (met by the reference row).

Use

aether run qwen3-1.7b --prompt "Hello"
import Aether

let aether = try Aether()
let chat = try await aether.chat("qwen3-1.7b")
let reply = try await chat.respond(to: "Hello")
print(reply.text)
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for aether-models/qwen3-1.7b

Finetuned
Qwen/Qwen3-1.7B
Finetuned
(1205)
this model