invoke-ai
[bug]: Cannot install Qwen-Image from huggingface
Is there an existing issue for this problem?
- [x] I have searched the existing issues
Install method
Invoke's Launcher
Operating system
Linux
GPU vendor
AMD (ROCm)
GPU model
RX 7900 XTX
GPU VRAM
24
Version number
v6.3.0rc1
Browser
Chrome Version 138.0.7204.183 (Official Build) (64-bit)
System Information
{ "version": "6.3.0rc1", "dependencies": { "absl-py" : "2.3.1", "accelerate" : "1.8.1", "addict" : "2.4.0", "aenum" : "3.1.16", "aiofiles" : "23.2.1", "aiohappyeyeballs" : "2.6.1", "aiohttp" : "3.12.14", "aiosignal" : "1.4.0", "altair" : "5.5.0", "annotated-types" : "0.7.0", "antlr4-python3-runtime" : "4.9.3", "anyio" : "4.9.0", "astroid" : "3.3.11", "astunparse" : "1.6.3", "attrs" : "25.3.0", "basicsr" : "1.4.2", "bidict" : "0.23.1", "bitsandbytes" : "0.46.1", "blake3" : "1.0.5", "blendmodes" : "2024", "certifi" : "2025.7.14", "cffi" : "1.17.1", "charset-normalizer" : "3.4.2", "clean-fid" : "0.1.35", "click" : "8.2.1", "clip" : "1.0", "cloudpickle" : "3.1.1", "coloredlogs" : "15.0.1", "compel" : "2.1.1", "contourpy" : "1.3.2", "CUDA" : "N/A", "cycler" : "0.12.1", "dctorch" : "0.1.2", "Deprecated" : "1.2.18", "deprecation" : "2.1.0", "diffusers" : "0.33.0", "dill" : "0.4.0", "diskcache" : "5.6.3", "dnspython" : "2.7.0", "dynamicprompts" : "0.31.0", "einops" : "0.4.1", "facexlib" : "0.3.0", "fastapi" : "0.116.1", "fastapi-events" : "0.12.2", "fasteners" : "0.19", "ffmpy" : "0.6.0", "filelock" : "3.18.0", "filetype" : "1.2.0", "filterpy" : "1.4.5", "flatbuffers" : "25.2.10", "fonttools" : "4.59.0", "frozenlist" : "1.7.0", "fsspec" : "2025.7.0", "ftfy" : "6.3.1", "future" : "1.0.0", "gfpgan" : "1.3.8", "gguf" : "0.17.1", "gitdb" : "4.0.12", "GitPython" : "3.1.44", "gradio" : "3.43.2", "gradio_client" : "0.5.0", "gradio_imageslider" : "0.0.20", "gradio_rangeslider" : "0.0.6", "greenlet" : "3.2.3", "grpcio" : "1.74.0", "h11" : "0.14.0", "hf-xet" : "1.1.6rc2", "httpcore" : "0.17.3", "httptools" : "0.6.4", "httpx" : "0.24.1", "huggingface-hub" : "0.33.1", "humanfriendly" : "10.0", "idna" : "3.10", "imageio" : "2.37.0", "importlib_metadata" : "8.7.0", "importlib_resources" : "6.5.2", "inflection" : "0.5.1", "invisible-watermark" : "0.2.0", "InvokeAI" : "6.3.0rc1", "isort" : "6.0.1", "jax" : "0.7.0", "jaxlib" : "0.7.0", "Jinja2" : "3.1.6", "jsonmerge" : "1.9.2", "jsonschema" : "4.24.1", "jsonschema-specifications" : "2025.4.1", "kiwisolver" : "1.4.8", "kornia" : "0.6.7", "lark" : "1.1.2", "lazy_loader" : "0.4", "lightning-utilities" : "0.14.3", "llvmlite" : "0.44.0", "lmdb" : "1.7.3", "loadimg" : "0.1.2", "Markdown" : "3.8.2", "markdown-it-py" : "3.0.0", "MarkupSafe" : "2.1.5", "matplotlib" : "3.10.3", "mccabe" : "0.7.0", "mdurl" : "0.1.2", "mediapipe" : "0.10.14", "ml_dtypes" : "0.5.1", "mpmath" : "1.3.0", "multidict" : "6.6.3", "narwhals" : "2.0.0", "networkx" : "3.5", "numba" : "0.61.2", "numexpr" : "2.10.2", "numpy" : "1.26.4", "omegaconf" : "2.2.3", "onnx" : "1.16.1", "onnxruntime" : "1.19.2", "open-clip-torch" : "2.20.0", "opencv-contrib-python" : "4.11.0.86", "opencv-contrib-python-headless": "4.9.0.80", "opencv-python" : "4.11.0.86", "opencv-python-headless" : "4.12.0.88", "opt_einsum" : "3.4.0", "optimum" : "1.26.1", "orjson" : "3.11.0", "packaging" : "25.0", "pandas" : "2.3.0", "patch-ng" : "1.18.1", "peft" : "0.15.2", "pi_heif" : "1.0.0", "picklescan" : "0.0.26", "piexif" : "1.1.3", "pillow" : "10.4.0", "pillow-avif-plugin" : "1.4.3", "pip" : "25.1.1", "platformdirs" : "4.3.8", "pooch" : "1.8.2", "prompt_toolkit" : "3.0.51", "propcache" : "0.3.2", "protobuf" : "4.25.3", "psutil" : "7.0.0", "py-cpuinfo" : "9.0.0", "pycparser" : "2.22", "pydantic" : "2.11.7", "pydantic-settings" : "2.10.1", "pydantic_core" : "2.33.2", "pydub" : "0.25.1", "Pygments" : "2.19.2", "pylint" : "3.3.7", "PyMatting" : "1.1.14", "pyparsing" : "3.1.4", "PyPatchMatch" : "1.0.2", "python-dateutil" : "2.9.0.post0", "python-dotenv" : "1.1.1", "python-engineio" : "4.12.2", "python-multipart" : "0.0.20", "python-socketio" : "5.13.0", "pytorch-lightning" : "1.9.4", "pytorch-triton-rocm" : "3.3.1", "pytz" : "2025.2", "PyWavelets" : "1.8.0", "PyYAML" : "6.0.2", "referencing" : "0.36.2", "regex" : "2024.11.6", "rembg" : "2.0.67", "requests" : "2.32.3", "resize-right" : "0.0.2", "rich" : "14.0.0", "rpds-py" : "0.26.0", "ruff" : "0.12.4", "safetensors" : "0.5.3", "scikit-image" : "0.21.0", "scipy" : "1.16.0", "seaborn" : "0.13.2", "seam-carving" : "1.1.0", "semantic-version" : "2.10.0", "semver" : "3.0.4", "sentencepiece" : "0.2.0", "setuptools" : "69.5.1", "shellingham" : "1.5.4", "simple-websocket" : "1.1.0", "six" : "1.17.0", "smmap" : "5.0.2", "sniffio" : "1.3.1", "sounddevice" : "0.5.2", "spandrel" : "0.4.1", "spandrel_extra_arches" : "0.1.1", "SQLAlchemy" : "2.0.41", "starlette" : "0.47.2", "sympy" : "1.14.0", "tb-nightly" : "2.21.0a20250725", "tensorboard-data-server" : "0.7.2", "tensordict" : "0.8.3", "tifffile" : "2025.6.11", "timm" : "1.0.16", "tokenizers" : "0.21.1", "tomesd" : "0.1.3", "toml" : "0.10.2", "tomlkit" : "0.12.0", "torch" : "2.7.1+rocm6.3", "torchdiffeq" : "0.2.3", "torchmetrics" : "1.7.4", "torchsde" : "0.2.6", "torchvision" : "0.22.1+rocm6.3", "tqdm" : "4.67.1", "trampoline" : "0.1.2", "transformers" : "4.53.0", "typer" : "0.16.0", "typing-inspection" : "0.4.1", "typing_extensions" : "4.12.2", "tzdata" : "2025.2", "ultralytics" : "8.3.40", "ultralytics-thop" : "2.0.14", "urllib3" : "1.26.19", "uvicorn" : "0.35.0", "uvloop" : "0.21.0", "voluptuous" : "0.15.2", "watchfiles" : "1.1.0", "wcwidth" : "0.2.13", "websockets" : "11.0.3", "Werkzeug" : "3.1.3", "wheel" : "0.45.1", "wrapt" : "1.17.2", "wsproto" : "1.2.0", "yapf" : "0.43.0", "yarl" : "1.20.1", "zipp" : "3.23.0" }, "config": { "schema_version": "4.0.2", "legacy_models_yaml_path": null, "host": "127.0.0.1", "port": 9090, "allow_origins": [""], "allow_credentials": false, "allow_methods": [""], "allow_headers": ["*"], "ssl_certfile": null, "ssl_keyfile": null, "log_tokenization": false, "patchmatch": true, "models_dir": "models", "convert_cache_dir": "models/.convert_cache", "download_cache_dir": "models/.download_cache", "legacy_conf_dir": "configs", "db_dir": "databases", "outputs_dir": "outputs", "custom_nodes_dir": "nodes", "style_presets_dir": "style_presets", "workflow_thumbnails_dir": "workflow_thumbnails", "log_handlers": ["console"], "log_format": "color", "log_level": "info", "log_sql": false, "log_level_network": "warning", "use_memory_db": false, "dev_reload": false, "profile_graphs": false, "profile_prefix": null, "profiles_dir": "profiles", "max_cache_ram_gb": 34, "max_cache_vram_gb": 12, "log_memory_usage": true, "device_working_mem_gb": 4, "enable_partial_loading": true, "keep_ram_copy_of_weights": false, "ram": null, "vram": null, "lazy_offload": true, "pytorch_cuda_alloc_conf": "backend:native", "device": "auto", "precision": "auto", "sequential_guidance": false, "attention_type": "auto", "attention_slice_size": "auto", "force_tiled_decode": false, "pil_compress_level": 1, "max_queue_size": 100, "clear_queue_on_startup": true, "allow_nodes": null, "deny_nodes": null, "node_cache_size": 512, "hashing_algorithm": "blake3_single", "remote_api_tokens": null, "scan_models_on_startup": false }, "set_config_fields": [ "max_cache_vram_gb", "enable_partial_loading", "max_queue_size", "max_cache_ram_gb", "legacy_models_yaml_path", "clear_queue_on_startup", "allow_origins", "pytorch_cuda_alloc_conf", "allow_credentials", "keep_ram_copy_of_weights", "device_working_mem_gb", "log_memory_usage" ] }
What happened
Cannot install Qwen/Qwen-Image https://huggingface.co/Qwen/Qwen-Image/tree/main Tried with url tab and huggingface tab in models section of Invokeai-web gui.
What you expected to happen
I expected qwen-image to be installed
How to reproduce the problem
Try to install the model as specified by InvokeAI
Additional context
No response
Discord username
No response
Could you please consider QWen to be in the invoke Family? Qwen is getting better and better. @psychedelicious
We are monitoring the ecosystem around it but as of this time it isn't a priority.
We need Qwen image, Qwen Image Edit and Qwen image controlnets support. What's stopping you guys?
We need Qwen image, Qwen Image Edit and Qwen image controlnets support. thanks！
well, I tried qwen in comfyui and I have to say that I'm impressed, it should be supported in invokeAI sooner or later
We need Qwen image, Qwen Image Edit and Qwen image controlnets support. thanks！
We know y'all want Qwen. We know that it is a great model.
Please just add a thumbs up to the first post in this GH issue. Every time you reply to this thread, it pings the devs and we stop whatever we are doing to go and read your message. It distracts us and delays implementing Qwen.
Please do not add comments or reply saying you want Qwen. Just add a thumbs up.
Not sure why the OP closed this. Reopened to track progress on this model
Does Qwen work on Nvidia CUDA GPUs with invoke ? Is it a general problem with the model loader or just an AMD/ROCm problem ?
Ist this still in focus? Given the quality of Qwen-Image, it seems pretty important.
I don't think they're interested. And that's not even to talk about the new Z-Image open models.
I don't think it's fair (or helpful) to say "They are not interested", or leave such Adobe comments. If the devs could just push a button and invoke Qwen support, they'd do it. But there is only a small number of devs working on Invoke after the core dev team left.
I've had discussions with some of the folks on Discord, and the sentiment back then was to get video support in Invoke going. I assume that's where some of the eng resources go. Personally, I disagree with this direction. Comfy already has excellent video support, and because you cannot edit videos the same way you can edit images, it's not building on Invoke's strengths.
Having said that, this was just the opinion of one dev. Nobody is preventing contributors from adding more image models. So don't give up hope, or start contributing even.
I agree with what you say about it being not helpful. However it's hard to remain dispassionate when this is such a great product and could build on its legacy. I also agree about the video dev. That seems to be really quite a divergence. Of course it's easy to talk when we're so far from the coal face.