GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
microsoft
Model provider
microsoft
Model tree
Base
this model
Modalities
Input
-
Output
-
Pricing
Dedicated Endpoints
View detailsSupported Functionality
Model APIs
Dedicated Endpoints
Container
More information
1:2 to 2:1 and resolutions up to 1440×1440.Tested environment: Python 3.12 · CUDA 12.6 · PyTorch 2.11.0+cu126 · TorchVision 0.26.0+cu126
bash
conda create -n lens python=3.12 -yconda activate lensuv pip install torch==2.11.0+cu126 torchvision==0.26.0+cu126 \--index-url https://download.pytorch.org/whl/cu126uv pip install -r requirements.txt
The default GPT-OSS encoder and FLUX.2 VAE are loaded from Hugging Face. Make sure your environment has access to any gated model repositories you use.
| Repo | Description | Steps | CFG |
|---|---|---|---|
microsoft/Lens | Default. RL-tuned for visual quality | 20 | 5.0 |
microsoft/Lens-Turbo | Distilled from the RL model for fast 4-step sampling | 4 | 1.0 |
Pick a variant by passing its repo id to --repo_id (CLI) or LensPipeline.from_pretrained(...) (Python).
Important: run from the cloned repo root so
from lens import LensPipelineresolves to this package — importinglensis what registersLensGptOssEncoder/LensTransformer2DModelwith thetransformersanddiffusersnamespaces thatmodel_index.jsonreferences.
Python API:
python
import torchfrom lens import LensPipelinepipe = LensPipeline.from_pretrained("microsoft/Lens", torch_dtype=torch.bfloat16).to("cuda")image = pipe(prompt="A cat holding a sign that says \"hello world\"",base_resolution=1440, aspect_ratio="1:1",num_inference_steps=20, guidance_scale=5.0,generator=torch.Generator("cuda").manual_seed(0),).images[0]image.save("lens.png")
To trade speed for VRAM, replace .to("cuda") with pipe.enable_model_cpu_offload().
CLI — basic usage:
bash
python inference.py \--repo_id "microsoft/Lens" \--prompt "A cinematic mountain lake at sunrise, soft mist, detailed reflections" \--base_resolution 1440 --aspect_ratio 1:1 \--steps 20 --cfg 5.0 --n 1 --seed 42 \--out ./outputs
Batch generation — join multiple prompts with |:
bash
python inference.py \--repo_id "microsoft/Lens" \--steps 20 --cfg 5.0 \--prompt "a red fox in snow|a glass greenhouse at night"
A100 / V100 (no MXFP4 kernels) — dequantize the GPT-OSS encoder to bf16:
bash
python inference.py \--repo_id "microsoft/Lens" \--steps 20 --cfg 5.0 \--prompt "a cat" \--disable_mxfp4 --offload
| Flag | Description | Default |
|---|---|---|
--repo_id | HF repo id (or local path) of the assembled Lens pipeline | microsoft/Lens |
--base_resolution | 1024 or 1440 | 1440 |
--aspect_ratio | 1:2, , , , , , , , |
bibtex
@article{zhao2026lens,title = {Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models},author = {Guo, Baining and Luo, Chong and Chen, Dong and Chen, Dongdong and Wei, Fangyun and Li, Ji and Bao, Jianmin and Zhang, Jiawei and Zhao, Jinjing and Shi, Lei and Yang, Qinhong and Zhang, Sirui and Wu, Xiuyu and Feng, Xuelu and Lu, Yan and Dong, Yanchen and Yue, Yang and Wang, Yitong and Chen, Yunuo and Liang, Zhiyang and Wan, Ziyu},journal = {arXiv preprint arXiv:2605.21573},year = {2026}}
The model is released for research purposes only and is not intended for product or service deployment. Responsible AI considerations were incorporated throughout the development process, including data selection, model training, and evaluation. The training data includes a combination of public, licensed, and internal datasets that were processed to remove clearly identifiable personal information and reduce harmful content where possible. However, as the data is largely sourced from web-scale collections, it may contain biases or uneven representation. As a result, the model may generate outputs that are inaccurate, biased, or inappropriate under certain prompts, including content that could be misleading or raise copyright or IP-related concerns. Given these limitations, the model should be used in controlled research settings, with appropriate human oversight. Downstream users are responsible for applying additional safeguards, such as content moderation, validation, and compliance checks, before using the model in broader applications.
This project does not collect any usage data. For more information, see the Microsoft Privacy Statement.
This project is released under the MIT License.
microsoft/Lens-Base| Supervised base model (no RL, no distillation) |
| 50 |
| 5.0 |
9:162:33:41:14:33:216:92:11:1 |
--steps | Number of denoising steps | 20 |
--cfg | Classifier-free guidance scale | 5.0 |
--n | Number of images per prompt | 1 |
--seed | Random seed (omit for non-deterministic) | — |
--out | Output directory | ./outputs |
--dtype | Compute dtype: bfloat16, float16, float32 | bfloat16 |
--disable_mxfp4 | Dequantize the GPT-OSS text encoder to --dtype (required on A100 / V100; Hopper+ keeps MXFP4 by default for less VRAM) | — |
--offload | Enable diffusers CPU offload (text_encoder->transformer->vae) to reduce peak VRAM | — |
--reasoner | Refine prompts with the loaded GPT-OSS encoder before generation | — |
--api_url / --api_key / --api_model | Use an OpenAI-compatible API for prompt refinement (takes precedence over --reasoner) | — |