---
title: SenseNova-U1.5-8B-MoT-8step-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit.md"
repository: mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit
last_updated: 2026-10-12
license: apache-2.0
model_type:
  - neo_chat
architectures:
  - NEOChatModel
base_model:
  - sensenova/SenseNova-U1.5-8B-MoT
base_model_relation: finetune
parameters: 3.9B
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 43
stars: 1
tags:
  - mlx
  - mlx-swift
  - image-generation
  - image-editing
  - any-to-any
  - "native multimodal"
  - distilled
---

# SenseNova-U1.5-8B-MoT-8step-4bit

> SenseNova-U1.5-8B-MoT-8step-4bit - mlx-community 在 ModelScope 开源的模型。SenseNova-U1.5-8B-MoT-8step-4bit

mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit 是 ModelScope 魔搭社区上的 3.9B 参数机器学习模型，采用 apache-2.0 许可，基于 sensenova/SenseNova-U1.5-8B-MoT 构建。

- **Repository**: mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit
- **License**: apache-2.0
- **Parameters**: 3.9B
- **Base model**: sensenova/SenseNova-U1.5-8B-MoT
- **Tags**: mlx, mlx-swift, image-generation, image-editing, any-to-any, native multimodal, distilled
- **Downloads**: 43
- **Stars**: 1
- **Last updated**: 2026-10-12

Source: https://www.modelscope.cn/models/mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit

---

# SenseNova-U1.5-8B-MoT-8step-4bit

Apple Silicon (MLX) artifact of **[SenseNova-U1.5-8B-MoT](https://huggingface.co/sensenova/SenseNova-U1.5-8B-MoT)** — SenseTime's unified T2I + editing + VQA flagship (NEO-unify Mixture-of-Transformers, pixel-space rectified flow, no VAE).

This artifact is the official **8-step distillation LoRA** (merged as above) plus **4-bit (group 64)** stream quantization.
Quantization: group size 64, 4-bit (mlx convention `quantization` block in config.json).

**Runtime:** [`sensenova-u1-swift`](https://github.com/xocialize/sensenova-u1-swift) (MLX-Swift). Weights are stored in the runtime's key layout (NHWC convs; quantized linears as `weight`/`scales`/`biases`) and load with **zero conversion transient** — peak load memory ≈ resident.

## Performance (M5 Max)

**peak 14.8 GB · 1024² in 3.2 s · 2048² in 15.8 s**

**Use it for:** the smallest, fastest tier for T2I and editing — fits 16 GB-class Macs. Note: 4-bit weight quantization on a diffusion path measurably perturbs the denoise trajectory: at a fixed seed it produces a *different but equally valid* image (cos ~0.92 vs bf16). Prefer the 8-bit tier when bf16-faithful output matters.

**Editing and VQA run on this tier too** (runtime ≥ 0.5.1, 2026-10-11): the distill adapter touches only the 294 generation-stream projections, so VQA uses the base tier's understanding stream unchanged (answers match token for token), and the 8-step cfg-free edit matches the 50-step base edit (a 2048² jacket edit in 15 s vs 169 s, 28.7 dB apart). Think mode stays a base-tier feature. Evidence: `Docs/receipts/8step-v2-ab/` and AB-R-0490 in the runtime repo.

## Quick start

```bash
git clone https://github.com/xocialize/sensenova-u1-swift && cd sensenova-u1-swift && swift build -c release
hf download mlx-community/SenseNova-U1.5-8B-MoT-8step-4bit --local-dir SenseNova-U1.5-8B-MoT-8step-4bit
.build/release/sensenova-cli --weights SenseNova-U1.5-8B-MoT-8step-4bit \
  --prompt "A cinematic mountain lake at sunrise, realistic photography." \
  --width 1024 --height 1024 --steps 8 --cfg 1.0 --out out.npy
```

The runtime also does instruction editing (`--edit-image`) and VQA (`--vqa`) on this tier, and `<think>` reasoning mode (`--think`) on the base tiers — see the [repository README](https://github.com/xocialize/sensenova-u1-swift) for the full surface and the parity report (component parity < 1e-4; e2e per-pass cosine 0.999+ vs the reference PyTorch implementation).

## Provenance & license

- **Upstream:** [sensenova/SenseNova-U1.5-8B-MoT](https://huggingface.co/sensenova/SenseNova-U1.5-8B-MoT) @ [`07d76f6`](https://huggingface.co/sensenova/SenseNova-U1.5-8B-MoT/commit/07d76f61474b9c6e6999e22c559314d3439b8c81), Apache-2.0, by [SenseTime / SenseNova](https://huggingface.co/sensenova) — [paper](https://arxiv.org/abs/2605.12500) · [reference implementation](https://github.com/OpenSenseNova/SenseNova-U1).
- This repository is a format conversion (official-LoRA merge + bf16 cast + 4-bit group-64 quantization) of the upstream checkpoint, redistributed under the same **Apache-2.0** license with modifications noted here. All credit for the model to the SenseNova team.
- `tokenizer.json` is generated from the upstream `vocab.json`/`merges.txt` (byte-identical tokenization, verified against reference ids).