whatmodel.sh / image / LongCat-Image
LongCat-Image
↗ share
+ contribute
LongCat-Image is an open-weights image model from Meituan, released December 2025. Its published size is 6B parameters. Licence: apache-2.0.
Model information sources
Release date ↗ · Weights licence ↗ · Parameter count ↗
The makers technical report (arXiv 2512.07584, CC BY 4.0) clarifies this is the core diffusion model: "With a core diffusion model of only 6B parameters"; text encoder/VAE not included.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of LongCat-Image, out of 197 with a score, then more from Meituan.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 86 |  HiDream-O1-Image-Dev | 54±5 | comparison only | ||
| 87 |  LLaDA-Image-Turbo | 54±3 | local | ||
| 88 | LongCat-Image | 53±4 | api $0.13/image +local | ||
| 89 |  Step1X-Edit v1.2 | 53±2 | local | ||
| 90 |  Kling Image 2.1 | 53±7 | api $0.014/image | ||
| more from Meituan | |||||
| 72 | LongCat-Image-Edit | 56±2 | api $0.15/image +local | ||
$ what it's good at
text to image▂▅▅▃▄▄█▃▁▂▁▁ #59 of 126 · 95% of its score · 7 tests
Beats FLUX.2 [klein] 9B, Qwen-Image, LLaDA-Image-Turbo here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| QC-T2I-Bench (English, HCQ total) | 83.9% ✎ | ▄····▄█▄▄█▄█ beats 1/10 | 68% |
| WISE (WiScore overall, GPT-4o judge) | 65.0% ✎ | ▄·▆▂▆█▄▆▂▄·▄ beats 20/22 | 23% |
| ChineseWord (LongCat-Image 8,105-character set, PPOCRv5) overall lab | 90.7% ✎ | ▄█·········▄ beats 3/3 | 3.3% |
| GenEval lab | 87.0% ✎ | ▇▃▅▇▁▄▅▃█▃▃▅ beats 35/39 | — |
| DPG-Bench lab | 86.8% ✎ | ▄·▂·▂▆█▄▂▄▂▂ beats 12/17 | — |
| CVTG-2K word accuracy lab | 86.6% ✎ | ▄▄▄█▄▄·▄···▄ beats 3/8 | — |
| WISE overall (model publisher's run) lab | 65.0% ✎ | no other model yet | — |
$ where to run it
Through an API
api · fal.ai · $0.13/image · fal.ai ↗ · price sourceUSD for one 1-megapixel output: fal bills this endpoint $0.13 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 text-to-image. API model: fal-ai/longcat-image (LongCat Image, 6B). Settings: 1 MP output, one image. Price basis: published per-megapixel list rate × 1 MP. Currency: USD. Not a Meituan direct tariff. fal.ai official model page, observed 2026-09-26.
Other ways to run it
- Run it as published · needs ~15 GB · from github.com/leejet
sd-cli --diffusion-model models/LongCat-Image-Q4_K_M.gguf --vae models/ae.safetensors --llm models/Qwen2.5-VL-7B-Instruct.Q8_0.gguf -p "a lovely cat holding a sign says 'longcat.cpp'" --cfg-scale 5.0 --sampling-method euler --flow-shift 3 -v --offload-to-cpu --diffusion-fa
Weights and documentation: huggingface.co/meituan-longcat ↗
$ how the score was worked out · QC-T2I-Bench: Scalable Questi… 68%, WISE: A World Knowledge-Infor… 23%, Meituan LongCat — LongCat-Ima… 3%
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 |
|---|---|---|
 QC-T2I-Bench: Scalable Question-Centric Text-to-Image Evaluation (Reliable Ranking, Fine-Grained Diagnosis, and Cost-Aware Routing) | 1 | 68% |
| WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation (arXiv v4) | 1 | 23% |
| Meituan LongCat — LongCat-Image technical report · lab-reported | 1 | 3% |
| prior · pull toward the middle | 5% | |
| weighted average | = 53.4 |
not counted (4)
| Meituan LongCat — LongCat-Image technical report | CVTG-2K word accuracy [correct] | 86.6% | fewer than three models have this test yet, or this source ran it for this model alone |
| Meituan LongCat — LongCat-Image technical report | DPG-Bench [correct] | 86.8% | fewer than three models have this test yet, or this source ran it for this model alone |
| Meituan LongCat — LongCat-Image technical report | GenEval [correct] | 87.0% | fewer than three models have this test yet, or this source ran it for this model alone |
| Meituan LongCat — LongCat-Image technical report | WISE overall (model publisher's run) [correct] | 65.0% | fewer than three models have this test yet, or this source ran it for this model alone |
$ 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