GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
mindlab-research
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
mindlab-research
Model Tree
zai-org/GLM-5.1
Input Modalities
Output Modalities
Supported Functionality
| Field | Value |
|---|---|
| Model family | Macaron A2UI |
| Variant | Tall |
| Release name | Macaron A2UI Venti |
| Release type | LoRA adapter |
| Foundation checkpoint | zai-org/GLM-5.1 |
| Target protocol | A2UI v0.8 |
| Output format | JSON object with text_response and a2ui fields |
| Training method | GRPO with LoRA |
| Library | PEFT / Transformers |
| Recommended dtype | bfloat16 |
| Tokenizer | Same as foundation checkpoint |
| Field | Value |
|---|---|
| LoRA rank | 16 |
| LoRA alpha | 32 |
| LoRA dropout | 0.0 |
| Target modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |
| LM head adapted | No |
| Training max response | 4096 |
| Variant | Release Name | Foundation Checkpoint | Release Type |
|---|---|---|---|
| Tall | Macaron A2UI Venti | Qwen/Qwen3-30B-A3B-Instruct-2507 | LoRA adapter |
| Grande | Macaron A2UI Grande | Qwen/Qwen3-235B-A22B-Instruct-2507 | LoRA adapter |
| Venti | Macaron A2UI Venti | GLM 5.1 | LoRA adapter |
You are currently viewing the Tall release.
This repository contains adapter weights only. Load the corresponding foundation checkpoint first, then attach this adapter with PEFT.
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torch base_model_id = "zai-org/GLM-5.1"adapter_id = "mindlab-research/Macaron-A2UI-Venti" tokenizer = AutoTokenizer.from_pretrained( base_model_id, trust_remote_code=True,) base_model = AutoModelForCausalLM.from_pretrained( base_model_id, torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True,) model = PeftModel.from_pretrained(base_model, adapter_id)model.eval() messages = [ { "role": "system", "content": "You are an A2UI v0.8 card generation model. Output exactly one valid A2UI JSON card." }, { "role": "user", "content": "<USER_CONTEXT_JSON>", },] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True,) inputs = tokenizer(text, return_tensors="pt").to(model.device) outputs = model.generate( **inputs, max_new_tokens=2048, do_sample=False,) response = tokenizer.decode( outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True,) print(response)Macaron A2UI Venti is trained to output:
{"text_response": "...", "a2ui": [...]};The a2ui field is expected to contain A2UI v0.8 messages such as beginRendering, surfaceUpdate, dataModelUpdate, or deleteSurface.
The model targets A2UI v0.8. Compatibility with later protocol revisions is not guaranteed without additional validation or fine-tuning.
We evaluate Macaron A2UI on internal A2UI v0.8 card-generation benchmarks and product-aligned task suites.
Public benchmark numbers and reproduction details are being standardized and will be added in a future revision of this model card.
At the moment, this repository should be interpreted as an adapter release first. Evaluation methodology, task definitions, and comparable public results are still being consolidated.
Macaron A2UI Venti is specialized for A2UI generation and is not intended as a general-purpose chat model.
Known limitations:
The adapter weights are released under MIT.
This adapter is trained on top of zai-org/GLM-5.1. Users are responsible for complying with both:
bibtex
@misc{kong2026macaron_a2ui, author = {Fancy Kong and Congjie Zheng and Murphy Zhuang and Rio Yang and Sueky Zhang and Hao Fu and Gene Jin and Andrew Chen and Pony Ma and {Mind Lab}}, title = {Macaron-A2UI: A Model for Generative UI in Personal Agent}, year = {2026}, howpublished = {Mind Lab: A Lab for Experiential Intelligence}, note = {https://macaron.im/mindlab/research/macaron-a2ui-generative-ui-personal-agent}}from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelimport torch base_model_id = "zai-org/GLM-5.1"adapter_id = "mindlab-research/Macaron-A2UI-Venti" tokenizer = AutoTokenizer.from_pretrained( base_model_id, trust_remote_code=True,) base_model = AutoModelForCausalLM.from_pretrained( base_model_id, torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True,) model = PeftModel.from_pretrained(base_model, adapter_id)model.eval() messages = [ { "role": "system", "content": "You are an A2UI v0.8 card generation model. Output exactly one valid A2UI JSON card." }, { "role": "user", "content": "<USER_CONTEXT_JSON>", },] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True,) inputs = tokenizer(text, return_tensors="pt").to(model.device) outputs = model.generate( **inputs, max_new_tokens=2048, do_sample=False,) response = tokenizer.decode( outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True,) print(response)@misc{kong2026macaron_a2ui, author = {Fancy Kong and Congjie Zheng and Murphy Zhuang and Rio Yang and Sueky Zhang and Hao Fu and Gene Jin and Andrew Chen and Pony Ma and {Mind Lab}}, title = {Macaron-A2UI: A Model for Generative UI in Personal Agent}, year = {2026}, howpublished = {Mind Lab: A Lab for Experiential Intelligence}, note = {https://macaron.im/mindlab/research/macaron-a2ui-generative-ui-personal-agent}}