whatmodel.sh / image / Emu3.5-Image
Emu3.5-Image
↗ share
+ contribute
Emu3.5-Image is an open-weights image model from BAAI, released October 2025. Its published size is 34.1B parameters. Licence: apache-2.0.
Model information sources
Release date ↗ · Weights licence ↗ · Parameter count ↗
Whole autoregressive model: HF Hub safetensors total for the official repo (pinned revision), 34,103,460,864 across its 15 model shards. The separate vision tokenizer repo is not included. Observed 2026-10-07.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of Emu3.5-Image, out of 197 with a score, then more from BAAI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 46 |  Recraft V4.1 Pro | 61±3 | api $0.21/image | ||
| 47 |  ERNIE-Image-Turbo | 61±3 | api $0.010/image +local | ||
| 48 | Emu3.5-Image | 60±5 | local | ||
| 49 | ERNIE-Image | 60±2 | api $0.030/image +local | ||
| 50 |  Kling Image O1 | 60±4 | api $0.028/image | ||
| more from BAAI | |||||
| 69 | Emu3.5 | 57±5 | local | ||
| 136 | OmniGen2 | 42±2 | api $0.15/image +local | ||
| 164 | Emu3 | 36±2 | local | ||
$ what it's good at
text to image▂▅▅▃▄▄█▃▁▂▁▁ #38 of 126 * · 90% of its score · 1 test
Beats Wan 2.7 Image Pro, Wan 2.7 Image here, though they rank higher overall. Trails ERNIE-Image here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| TIIF-Bench v3 (short prompts, TIIF-Evaluator) | 88.4% ✎ | ▂█····▄▄▄··▂ beats 9/11 | 90% |
$ where to run it
Other ways to run it
- Run it as published · needs ~84.5 GB, NVIDIA only · from huggingface.co/BAAI (official)
git clone https://github.com/baaivision/Emu3.5 && cd Emu3.5 && pip install -r requirements/transformers.txt && pip install flash_attn==2.8.3 --no-build-isolation && CUDA_VISIBLE_DEVICES=0 python inference.py --cfg configs/example_config_t2i.py && python src/utils/vis_proto.py --input outputs/emu3p5-image
Official weights and documentation: huggingface.co/BAAI ↗
$ how the score was worked out · TIIF-Bench: How Does Your T2I… 90%
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 |
|---|---|---|
 TIIF-Bench: How Does Your T2I Model Follow Your Instructions? (arXiv v3) | 1 | 90% |
| prior · pull toward the middle | 10% | |
| weighted average | = 60.4 |
$ 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