Downloads · 30 days
0
skillsafe-ai/qwen2.5-0.5b-instruct-q4f16
qwen2.5-0.5b-instruct-q4f16 is a text generation model from skillsafe-ai. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
Browser-ready mlc artifacts for text-generation, produced by SkillSafe's reproducible converter (models/ in skillsafe.ai) from a pinned upstream source. Every byte here is derivable from that source plus the recipe be…
Downloads · 30 days
0
Access
Public
Updated Sep 22, 2026
Repo size
283 MB
Likes
0
Public
Click a slice to open those files.
.bin278 MB · 94%
From the Hugging Face model README
Browser-ready mlc artifacts for text-generation, produced by SkillSafe's
reproducible converter (models/ in skillsafe.ai)
from a pinned upstream source. Every byte here is derivable from that source plus
the recipe below; nothing was edited by hand.
| Upstream | https://huggingface.co/mlc-ai/Qwen2.5-0.5B-Instruct-q4f16_1-MLC/tree/32ff081fe7e4dfe4ffb167b94c66fdf11e02b8ad |
| Upstream SHA-256 / commit | 32ff081fe7e4dfe4ffb167b94c66fdf11e02b8ad |
| Recipe | recipes/qwen2.5-0.5b-instruct-q4f16-prebuilt.yaml — sha256 c7a7d05a0de73a97fa5ab96bec6be4eea7ff118d745165b88ecedc21b30ab10c |
| Toolchain | Python 3.12.13, torch 2.10.0, onnx 1.23.0, onnxruntime 1.30.0 on Darwin 25.6.0 arm64 |
| Converted | 2026-09-22T03:19:15+00:00 |
| file | class | size | SHA-256 |
|---|---|---|---|
lib/Qwen2-0.5B-Instruct-q4f16_1-ctx4k_cs1k-webgpu.wasm | registry-shared | 4.78 MB | 25495a03bf7ac80a1925bc91e236c74c3ad52be89a9ed0d0eeb18e4a9d135234 |
merges.txt | bundle | 1.59 MB | 599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3 |
mlc-chat-config.json | bundle | 0.00 MB | 5439c03bdf4ee7cfe0f8a97a6588018b0144dc1280c71bbaea2853cdbab874b9 |
ndarray-cache.json | bundle | 0.10 MB | 23a54b5e9b271204561901b513c5b359498733429ee219ccc907eb1cd0790e3f |
params_shard_0.bin | registry | 64.91 MB | 9f309954d310dc63adfaf3ef6aa987c681b8aa6d1b9686aa2525b454b0d058d5 |
params_shard_1.bin | registry | 31.69 MB | 6d174758dd299d9ef4222b1ad4283be832ebd43853951da25b50501ab1b75ba7 |
params_shard_2.bin | registry | 31.95 MB | 83e0b530bf5c44cbead1a6c220af81040a975f7c81fb708977a02e3ac8d7ffa7 |
params_shard_3.bin | registry | 31.52 MB | 5ff16197c197d8783d398d0c35fa9641e606e6e2dc1d53b9f26a0c9c17a97921 |
params_shard_4.bin | registry | 31.49 MB | a7a3d2b02aa9258154f250a714d1743672e423c5c7c8e5c5eefcb9bf337aa0fd |
params_shard_5.bin | registry | 27.86 MB | 19dfd7a3064b84082915575c0e5a57fc1cd7108828e1ce9fbbbaf9db4b63b9af |
params_shard_6.bin | registry | 31.75 MB | 192576d43956aa977ec60848b8a7fc8483b5fe38ca9669aa9a3ca2ba795a7a33 |
params_shard_7.bin | registry | 13.93 MB | 1ee25c2a41dad6833e000b7e3ec13a5a1761c32ffbed0ad8a98a6ad313338dc0 |
tensor-cache.json | bundle | 0.10 MB | 23a54b5e9b271204561901b513c5b359498733429ee219ccc907eb1cd0790e3f |
tokenizer.json | bundle | 6.71 MB | c0382117ea329cdf097041132f6d735924b697924d6f6fc3945713e96ce87539 |
tokenizer_config.json | bundle | 0.01 MB | 5214600ee45ca2f887ce2eede8910378a0111ea99d657428bcbce94778e65a92 |
vocab.json | bundle | 2.65 MB | ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910 |
registry files are parameter files served from models.skillsafe.ai once vetted;
bundle files ship inside an app; registry-shared is a runtime library reused by
every model of the same architecture.
Shard sizes and md5s match ndarray-cache.json; the implied parameter count matches the checkpoint (494,032,768 parameters, 4.50 bits/param); the parameter table embedded in the WebGPU model library matches all 267 tensors of these weights.
import { CreateMLCEngine } from "@mlc-ai/web-llm";
const engine = await CreateMLCEngine("qwen2.5-0.5b-instruct-q4f16", {
appConfig: { model_list: [{
model: "https://huggingface.co/skillsafe-ai/qwen2.5-0.5b-instruct-q4f16",
model_id: "qwen2.5-0.5b-instruct-q4f16",
model_lib: "https://huggingface.co/skillsafe-ai/qwen2.5-0.5b-instruct-q4f16/resolve/main/lib/Qwen2-0.5B-Instruct-q4f16_1-ctx4k_cs1k-webgpu.wasm",
overrides: { context_window_size: 4096, prefill_chunk_size: 1024 },
}] },
});
The model library is compiled for context window 4096 / prefill chunk 1024; mlc-chat-config.json says 32768 / 2048, so set the overrides above.
Qwen2.5-0.5B-Instruct: Copyright 2024 Alibaba Cloud. Apache License 2.0. https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct — q4f16_1 weights published by MLC (https://huggingface.co/mlc-ai/Qwen2.5-0.5B-Instruct-q4f16_1-MLC).
Licence: Apache-2.0 — notice: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/7ae557604adf67be50417f59c2c2f167def9a775/LICENSE. The conversion recipe and this model card are part of the SkillSafe repository and carry its licence; the weights remain under the upstream licence above.
The full manifest.json in this repo records the recipe, sources, toolchain
(including the uv.lock hash) and per-file verification numbers.