whatmodel.sh / image / Qwen-Image-Edit-2509
Qwen-Image-Edit-2509
↗ share
+ contribute
Qwen-Image-Edit-2509 is an open-weights image model from Alibaba, released September 2025. Its published size is 20.43B parameters. Licence: apache-2.0.
Model information sources
Release date ↗ · Weights licence ↗ · Parameter count ↗
MMDiT transformer only: HF Hub safetensors total for the official repo (pinned revision), 20,430,401,088 = the sum of the transformer/ shard headers. The Qwen2.5-VL text encoder and the VAE are not included (convention of qwen-image = 20). The card says the model is 'Built upon our 20B Qwen-Image model'. Observed 2026-10-07.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of Qwen-Image-Edit-2509, out of 197 with a score, then more from Alibaba.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 57 |  GPT Image 1 Mini | 59±2 | api $0.011/image | ||
| 58 |  Krea 2 Large | 59±3 | api $0.060/image | ||
| 59 | Qwen-Image-Edit-2509 | 59±3 | api $0.030/image +local | ||
| 60 |  HiDream-O1-Image | 58±2 | api $0.010/image +local | ||
| 61 |  FLUX.2 [dev] | 58±1 | api $0.046/image +local | ||
| more from Alibaba | |||||
| 11 | Qwen-Image-3.0-Pro | 75±2 | api $0.040/image | ||
| 18 | Qwen-Image-2.1 | 69±2 | local | ||
| 20 | Qwen-Image-2.0-Pro (2026-06-22) | 67±2 | api $0.075/image | ||
$ what it's good at
image to image / editing▁▁·▃▂▃█▃▃▄▁▁ #31 of 62 · 74% of its score · 4 tests
Beats FLUX.2 [flex], GPT Image 1, Imagen 4 here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CEdit-Bench-CN overall (G_O, GPT-4o judge) | 73.7% ✎ | ▃·········█▂ beats 3/7 | 32% |
| CEdit-Bench-EN overall (G_O, GPT-4o judge) | 74.8% ✎ | ▄·▄···▄██··▄ beats 3/7 | 20% |
| GEditBench v2 (overall Elo) | 1014.0 ✎ | ▃▂····▂█▂▃▂▃ beats 8/14 | 17% |
| OmniContext average (GPT-4.1 judge) | 76.9% ✎ | ▃▃▃··▃·····█ beats 4/6 | 5.4% |
other tests2 not in a task · 23% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| RealPref-Bench image editing (human user-study score) | 86.7% ✎ | █▄▄····▄···▄ beats 5/5 | 16% |
| ViGoR-Bench zero-shot visual reasoning, image models (Result Avg) | 32.2% ✎ | ▂···▂·▂█▆▅·▂ beats 9/15 | 6.8% |
$ where to run it
Through an API
api · fal.ai · $0.030/image · fal.ai ↗ · price sourceUSD for one 1-megapixel output: fal bills this endpoint $0.03 per megapixel (price = rate × 1 MP; fal's rounding for 1024×1024 ≈ 1.05 MP is not stated; larger outputs cost proportionally more). Product: fal.ai serverless image editing. API model: fal-ai/qwen-image-edit-2509 (fal describes it as Qwen's Image Editing Plus model also known as Qwen-Image-Edit-2509). Settings: one input image, 1 MP output, one image. Price basis: published per-megapixel list rate × 1 MP. Alibaba Model Studio's own qwen-image-edit/-plus API models are not used because Alibaba does not state which open checkpoint they correspond to. Currency: USD. Not an Alibaba direct tariff. fal.ai official model page, observed 2026-09-26.
Other ways to run it
- Run it as published · needs ~26.2 GB · from github.com/leejet
sd-cli --diffusion-model models/Qwen-Image-Edit-2509-Q4_K_S.gguf --vae models/qwen_image_vae.safetensors --llm models/Qwen2.5-VL-7B-Instruct.Q8_0.gguf --llm_vision models/Qwen2.5-VL-7B-Instruct.mmproj-Q8_0.gguf --cfg-scale 2.5 --sampling-method euler -v --offload-to-cpu --diffusion-fa --flow-shift 3 -r input.png -p "change 'flux.cpp' to 'Qwen Image Edit 2509'"
Official weights and documentation: huggingface.co/Qwen ↗
$ how the score was worked out · Meituan LongCat — LongCat-Ima… 52%, GEditBench v2: A Human-Aligne… 17%, HP-Edit: A Human-Preference P… 16% and 2 more
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 Meituan LongCat — LongCat-Image technical report | 2 | 52% |
| GEditBench v2: A Human-Aligned Benchmark for General Image Editing | 1 | 17% |
| HP-Edit: A Human-Preference Post-Training Framework for Image Editing (arXiv:2604.19406v1) | 1 | 16% |
| ViGoR-Bench: How Far Are Visual Generative Models From Zero-Shot Visual Reasoners? (arXiv:2603.25823v1) | 1 | 7% |
| BAAI — OmniGen2 technical report | 1 | 5% |
| prior · pull toward the middle | 3% | |
| weighted average | = 58.6 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json