GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
alibaba-multimodal-industrial-ai
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
alibaba-multimodal-industrial-ai
Model Tree
Qwen/Qwen3.5-35B-A3B-Base
Input Modalities
Output Modalities
Supported Functionality
42络钼 → 42CrMo), and truncated standard codes (16674 → GB/T 16674), while intercepting contradictory geometric specifications before catalog dispatch.Think): Activates deep multi-step engineering reasoning for complex analytical and constraint-satisfaction tasks.No-Think): Strips intermediate CoT for latency-critical production environments (under 2s SLA).Qwen3_5MoeForConditionalGeneration)We recommend using transformers>=5.2.0 with native Qwen3.5 MoE support.
bash
pip install "transformers>=5.2.0" accelerateNo-Think, Recommended for Low-Latency Query Structuring)python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "alibaba-multimodal-industrial-ai/IndustryLLM" tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, device_map="auto",) # Example: Resolving colloquial buyer inquiry with conflicting specsprompt = "帮我找 42络钼 螺栓 16674 M10*35 但要50长" messages = [ {"role": "system", "content": "你是一位专业的工业品采销专家。请将买家的采购意图进行参数结构化归一化，指出潜在的规格冲突或缺失项。"}, {"role": "user", "content": prompt}] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, enable_thinking=False, # Direct response for fast catalog integration (<2s) tokenize=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(): output_ids = model.generate(**inputs, max_new_tokens=512, temperature=0.2) completion_ids = output_ids[0, inputs["input_ids"].shape[-1] :]print(tokenizer.decode(completion_ids, skip_special_tokens=True))Think, for Deep Engineering Analysis)python
# Pass enable_thinking=True to activate test-time reasoning tracesinputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, enable_thinking=True, # Activates chain-of-thought tokenize=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(): output_ids = model.generate(**inputs, max_new_tokens=2048, temperature=0.6) completion_ids = output_ids[0, inputs["input_ids"].shape[-1] :]print(tokenizer.decode(completion_ids, skip_special_tokens=True))As formalized in our technical report, category relevance is not product eligibility:
Satisfied, Unknown, Violated): unverified product claims or missing supplier evidence strictly evaluate to unknown rather than satisfied ("unknown is not satisfied").If you find IndustryLLM useful for your research or enterprise applications, please cite our technical report:
bibtex
@article{industryllm2026, title = {IndustryLLM: Failure-Driven LLM Training for Industrial Procurement}, author = {Multimodal and Industrial AI Team, Alibaba}, journal = {Technical Report}, year = {2026}, url = {https://huggingface.co/alibaba-multimodal-industrial-ai/IndustryLLM/resolve/main/IndustryLLM_paper.pdf}}Please also cite the upstream Qwen3.5 foundation model:
bibtex
@misc{qwen3_5, title = {Qwen3.5: Towards Native Multimodal Agents}, author = {Qwen Team}, year = {2026}, month = {February}, url = {[https://qwen.ai/blog?id=qwen3.5](https://qwen.ai/blog?id=qwen3.5)}}import torchfrom transformers import AutoModelForCausalLM, AutoTokenizer model_id = "alibaba-multimodal-industrial-ai/IndustryLLM" tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.bfloat16, device_map="auto",) # Example: Resolving colloquial buyer inquiry with conflicting specsprompt = "帮我找 42络钼 螺栓 16674 M10*35 但要50长" messages = [ {"role": "system", "content": "你是一位专业的工业品采销专家。请将买家的采购意图进行参数结构化归一化，指出潜在的规格冲突或缺失项。"}, {"role": "user", "content": prompt}] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, enable_thinking=False, # Direct response for fast catalog integration (<2s) tokenize=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(): output_ids = model.generate(**inputs, max_new_tokens=512, temperature=0.2) completion_ids = output_ids[0, inputs["input_ids"].shape[-1] :]print(tokenizer.decode(completion_ids, skip_special_tokens=True))# Pass enable_thinking=True to activate test-time reasoning tracesinputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, enable_thinking=True, # Activates chain-of-thought tokenize=True, return_dict=True, return_tensors="pt",).to(model.device) with torch.inference_mode(): output_ids = model.generate(**inputs, max_new_tokens=2048, temperature=0.6) completion_ids = output_ids[0, inputs["input_ids"].shape[-1] :]print(tokenizer.decode(completion_ids, skip_special_tokens=True))@article{industryllm2026, title = {IndustryLLM: Failure-Driven LLM Training for Industrial Procurement}, author = {Multimodal and Industrial AI Team, Alibaba}, journal = {Technical Report}, year = {2026}, url = {https://huggingface.co/alibaba-multimodal-industrial-ai/IndustryLLM/resolve/main/IndustryLLM_paper.pdf}}@misc{qwen3_5, title = {Qwen3.5: Towards Native Multimodal Agents}, author = {Qwen Team}, year = {2026}, month = {February}, url = {[https://qwen.ai/blog?id=qwen3.5](https://qwen.ai/blog?id=qwen3.5)}}