GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AIMH
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
AIMH
Model Tree
Qwen/Qwen2.5-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
Research on synthetic mental-health dialogue: generating therapist-side turns in CBT-style counseling conversations, studying privacy-preserving synthetic data, and benchmarking counseling-oriented language models. For the full questionnaire-conditioned, dual-agent (therapist + client) generation pipeline, see the code repository.
As a starting point for further research fine-tuning, or as a component in supervised, human-in-the-loop training and education settings (e.g., clinician/student practice simulations) under appropriate oversight and ethics approval.
This model must not be used to:
google/gemma-3-27b-it) used to generate the training corpus.Keep a qualified human professional in the loop for any applied use, validate on your own population, obtain your own ethics approval before any study involving people, and add explicit safety guardrails and crisis-resource handling in any interactive system. See the project ETHICS statement.
python
from transformers import AutoModelForCausalLM, AutoTokenizer model_id = "AIMH/SQPsych-8b-gemma-Qwen_no_questionnaire"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, dtype="bfloat16", device_map="auto") messages = [ {"role": "system", "content": "You are an empathetic therapist conducting a CBT-informed session."}, {"role": "user", "content": "I've felt down and unmotivated for weeks and I don't know why."},]inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, return_tensors="pt").to(model.device) outputs = model.generate(inputs, max_new_tokens=512, temperature=0.7, do_sample=True)print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))Serve with vLLM (OpenAI-compatible API):
bash
vllm serve "AIMH/SQPsych-8b-gemma-Qwen_no_questionnaire"AIMH/SQPsychConv_gemma_no_questionnaire, synthetic therapist-client conversations generated by google/gemma-3-27b-it from the de-identified structured profiles of the cohort of Kircher et al. (2019), without questionnaire-score conditioning (the no-questionnaire ablation). The instruction-formatted split used for training is AIMH/SQPsychConv_gemma_no_questionnaire_finetune.
The model weights derive from Qwen2.5-7B-Instruct (Apache-2.0). The training data was generated by google/gemma-3-27b-it, so Google's Gemma Terms of Use may apply to the synthetic data and to downstream use; review them before redistribution. The source structured data is de-identified and pre-anonymized, and the released conversations are synthetic and contain no personally identifiable information. Released for research only.
from transformers import AutoModelForCausalLM, AutoTokenizer model_id = "AIMH/SQPsych-8b-gemma-Qwen_no_questionnaire"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id, dtype="bfloat16", device_map="auto") messages = [ {"role": "system", "content": "You are an empathetic therapist conducting a CBT-informed session."}, {"role": "user", "content": "I've felt down and unmotivated for weeks and I don't know why."},]inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, return_tensors="pt").to(model.device) outputs = model.generate(inputs, max_new_tokens=512, temperature=0.7, do_sample=True)print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))