Veizik for Apple SiliconAvailable now
Local LLMs on your Mac, on supported Apple Silicon generations: M1, M2, M3, M4.
Whose work, which task.
Developers who run open language models on a Mac and want one signed binary instead of a framework stack: a quick chat, a one-shot run, or a local HTTP endpoint with OpenAI-style request and response shapes for their own tools.
What comes back.
$ veizik run qwen2.5-1.5b-instruct "In two sentences, what is a safetensors checkpoint?" --tokens 96
A safetensors checkpoint is a file that contains the weights and biases of a neural network model, which are used to train the model and generate the model's output. The safetensors format is designed to be more efficient and portable than the original TensorFlow checkpoint format, and it allows for better compression and storage of the model's weights. It is commonly used in the field of machine learning and artificial intelligence to save and load neural network models.
Mac Studio (M1 Ultra), the shipped 2026.10.01 disk image, 2026-10-01, greedy decoding, first run of a session: 1.22 s from the command to the full answer. Printed as generated; the model's description of the format is its own.
Where it runs, and what.
- Apple Silicon Mac (M1 or newer), macOS 13 or later, arm64 only. Intel Macs are not supported.
- Tested on macOS 15.5 (M2 Pro), 15.7.7 (M4 Pro), 26.5 (M1 Max) and 26.5.1 (M1 Ultra).
8 catalogue ids in the download. The measured table, with device memory per model and the build it was taken on, is on the hardware page.
| Model | Id | Checkpoint | Answers |
|---|---|---|---|
| Qwen2.5-0.5B | qwen2.5-0.5b | Qwen/Qwen2.5-0.5B | completion |
| Qwen2.5-1.5B | qwen2.5-1.5b | Qwen/Qwen2.5-1.5B | completion |
| Qwen2.5-7B | qwen2.5-7b | Qwen/Qwen2.5-7B | completion |
| Qwen2.5-0.5B-Instruct | qwen2.5-0.5b-instruct | Qwen/Qwen2.5-0.5B-Instruct | chat and completion |
| Qwen2.5-1.5B-Instruct | qwen2.5-1.5b-instruct | Qwen/Qwen2.5-1.5B-Instruct | chat and completion |
| Qwen2.5-7B-Instruct | qwen2.5-7b-instruct | Qwen/Qwen2.5-7B-Instruct | chat and completion |
| Qwen3-1.7B | qwen3-1.7b | Qwen/Qwen3-1.7B | chat and completion |
| Qwen3-1.7B-Base | qwen3-1.7b-base | Qwen/Qwen3-1.7B-Base | completion |
Veizik for Apple Silicon
Download, then the minimal commands.
Image rendered by Veizik 1.0.0 (sealed bundle, linux x86_64) · SD3.5-medium · NVIDIA RTX 3090 · Powered by Stability AI
From the package to a first answer.
Register this device. It opens your browser so you can approve the machine from your account, and prints the link as well in case it cannot. One activation per device -- not per model. Personal, educational, research and evaluation use at no charge, including internal evaluation inside a company; commercial use needs a separate licence from LinkPick; publishing benchmarks is permitted (LICENSE sections 2 and 4 in the disk image).
veizik-metal-2026.10.02.dmg · 2.94 MB · Apple disk image (.dmg), signed and notarised
sha256 3c288836d950847b46b53293770b33d06e8e6e2d1c82da448c639de0fcadc1ec
-
Install
on your machineOne signed package. One command in a shell, and it is ready.
./install.sh
-
Create an account
needs the networkAn email link, Google or GitHub. Self-service, nobody in the loop.
veizik.com/signin
-
Activate this device
needs the networkThe command opens your browser — and prints the link too, in case it cannot. You approve there, and the machine receives a licence bound to it.
veizik activate
-
Pull a checkpoint
needs the networkOne command downloads a supported safetensors checkpoint into the local cache and makes it ready to run. A gated repository needs a token stored with veizik hf login; the licence is accepted at the publisher.
veizik pull
-
Run
on your machineGeneration happens on the device. Your prompt, the weights and the output stay on the machine; the first run of a session fetches a short-lived licence lease and the runs that reuse it need no connection.
veizik run
3 of the 5 steps reach the network. Generation is not one of them.
$ shasum -a 256 veizik-metal-2026.10.02.dmg # 3c288836d950847b46b53293770b33d06e8e6e2d1c82da448c639de0fcadc1ec $ ./install.sh $ veizik activate $ veizik pull qwen2.5-7b $ veizik chat qwen2.5-7b $ veizik run qwen2.5-7b "The capital of France is" $ veizik serve qwen2.5-7b --port 8000
Measured on this build.
The whole installation: the CLI, four model cores and the signed core-id table. Summed from the published disk image, which is what the installer copies.
Standard checkpoints are prepared in memory.
4 of them answer a chat request as well as a completion.
Every row, and how to reproduce it.
Every published row names the device, the model, the token counts, the other engine's version and settings, how each side was timed, and the spread. Rows taken on earlier builds of the same cores are marked. The board →
$ veizik bench qwen2.5-7b
What it does not do, and where to go.
- veizik serve needs a model name in this build; streaming and request fields such as temperature, top_p, seed, tools and response_format are rejected with an explicit error.
- Qwen3-1.7B returns its reasoning text inside the answer; add /no_think to the prompt for a direct reply.
- Not supported: Intel Macs, Windows, Linux. Activation needs a browser on a device you control.
- Where another engine is ahead on this build is on the benchmarks board, row by row.
Troubleshooting: the docs and the release repository's issues. Terms: Personal, educational, research and evaluation use at no charge, including internal evaluation inside a company; commercial use needs a separate licence from LinkPick; publishing benchmarks is permitted (LICENSE sections 2 and 4 in the disk image). Model licences are the publishers' own — which models you may use commercially.