feat(models): add support to qwen-image

backend/python/diffusers/backend.py

@@ -65,6 +65,19 @@ def sc(self, clip_input, images): return images, [False for i in images]
     UniPCMultistepScheduler,
 )
 
+def is_float(s):
+    try:
+        float(s)
+        return True
+    except ValueError:
+        return False
+
+def is_int(s):
+    try:
+        int(s)
+        return True
+    except ValueError:
+        return False
 
 # The scheduler list mapping was taken from here: https://github.com/neggles/animatediff-cli/blob/6f336f5f4b5e38e85d7f06f1744ef42d0a45f2a7/src/animatediff/schedulers.py#L39
 # Credits to https://github.com/neggles
@@ -169,8 +182,24 @@ def LoadModel(self, request, context):
                 if ":" not in opt:
                     continue
                 key, value = opt.split(":")
+                # if value is a number, convert it to the appropriate type
+                if is_float(value):
+                    value = float(value)
+                elif is_int(value):
+                    value = int(value)
                 self.options[key] = value
 
+            # From options, extract if present "torch_dtype" and set it to the appropriate type
+            if "torch_dtype" in self.options:
+                if self.options["torch_dtype"] == "fp16":
+                    torchType = torch.float16
+                elif self.options["torch_dtype"] == "bf16":
+                    torchType = torch.bfloat16
+                elif self.options["torch_dtype"] == "fp32":
+                    torchType = torch.float32
+                # remove it from options
+                del self.options["torch_dtype"]
+
             print(f"Options: {self.options}", file=sys.stderr)
 
             local = False

gallery/index.yaml

@@ -1,4 +1,17 @@
 ---
+- &qwenimage
+  name: "qwen-image"
+  url: "github:mudler/LocalAI/gallery/qwen-image.yaml@master"
+  urls:
+    - https://huggingface.co/Qwen/Qwen-Image
+  icon: https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-Image/qwen_image_logo.png
+  license: apache-2.0
+  tags:
+    - qwen-image
+    - gpu
+    - text-to-image
+  description: |
+    We are thrilled to release Qwen-Image, an image generation foundation model in the Qwen series that achieves significant advances in complex text rendering and precise image editing. Experiments show strong general capabilities in both image generation and editing, with exceptional performance in text rendering, especially for Chinese.
 - &gptoss
   name: "gpt-oss-20b"
   url: "github:mudler/LocalAI/gallery/harmony.yaml@master"
@@ -2023,7 +2036,7 @@
     - gemma3
     - gemma-3
   overrides:
     #mmproj: gemma-3-27b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-27b-it-Q4_K_M.gguf
   files:
@@ -2041,7 +2054,7 @@
   description: |
     google/gemma-3-12b-it is an open-source, state-of-the-art, lightweight, multimodal model built from the same research and technology used to create the Gemini models. It is capable of handling text and image input and generating text output. It has a large context window of 128K tokens and supports over 140 languages. The 12B variant has been fine-tuned using the instruction-tuning approach. Gemma 3 models are suitable for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes them deployable in environments with limited resources such as laptops, desktops, or your own cloud infrastructure.
   overrides:
     #mmproj: gemma-3-12b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-12b-it-Q4_K_M.gguf
   files:
@@ -2059,7 +2072,7 @@
   description: |
     Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. Gemma 3 models are multimodal, handling text and image input and generating text output, with open weights for both pre-trained variants and instruction-tuned variants. Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions. Gemma 3 models are well-suited for a variety of text generation and image understanding tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as laptops, desktops or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. Gemma-3-4b-it is a 4 billion parameter model.
   overrides:
     #mmproj: gemma-3-4b-it-mmproj-f16.gguf
     parameters:
       model: gemma-3-4b-it-Q4_K_M.gguf
   files:
@@ -5525,7 +5538,7 @@
       sha256: 2756551de7d8ff7093c2c5eec1cd00f1868bc128433af53f5a8d434091d4eb5a
       uri: huggingface://Triangle104/Nano_Imp_1B-Q8_0-GGUF/nano_imp_1b-q8_0.gguf
 - &qwen25
   name: "qwen2.5-14b-instruct" ## Qwen2.5
   icon: https://avatars.githubusercontent.com/u/141221163
   url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
   license: apache-2.0
@@ -7486,7 +7499,7 @@
       sha256: 0fec82625f74a9a340837de7af287b1d9042e5aeb70cda2621426db99958b0af
       uri: huggingface://bartowski/Chuluun-Qwen2.5-72B-v0.08-GGUF/Chuluun-Qwen2.5-72B-v0.08-Q4_K_M.gguf
 - &smollm
   url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## SmolLM
   name: "smollm-1.7b-instruct"
   icon: https://huggingface.co/datasets/HuggingFaceTB/images/resolve/main/banner_smol.png
   tags:

gallery/qwen-image.yaml

@@ -0,0 +1,19 @@
+---
+name: "flux"
+
+config_file: |
+    backend: diffusers
+    cfg_scale: 0
+    diffusers:
+      cuda: true
+      enable_parameters: num_inference_steps
+      pipeline_type: DiffusionPipeline
+    f16: true
+    low_vram: true
+    name: qwen-image
+    parameters:
+      model: Qwen/Qwen-Image
+    step: 50
+    options:
+      - true_cfg_scale:4.0
+      - torch_dtype:bf16