GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
15MAI
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
15MAI
Model Tree
Input Modalities
Output Modalities
Supported Functionality
The model is built upon a decoder-only Transformer architecture optimized for speed and precise token classification:
The router classifies incoming developer prompts into five core technical domains:
code_python: Python scripts, data processing libraries (Pandas, NumPy), and asynchronous backend frameworks (FastAPI).code_bash: Command-line operations, log analysis (grep, awk), file searches, and system utilities.json_format: JSON payload manipulation, schema validation, and configuration parsing.sql_query: Database operations, PostgreSQL queries, joins, and indexing statements.system_control: Container management (Docker), resource monitoring, and service lifecycle commands.You can load and run the model locally using the Hugging Face transformers library:
python
import torchfrom transformers import GPT2LMHeadModel, PreTrainedTokenizerFast # Load model and tokenizertokenizer = PreTrainedTokenizerFast.from_pretrained("15MAI/15m-pico-router")model = GPT2LMHeadModel.from_pretrained("15MAI/15m-pico-router")model.eval() # Prepare structured inputprompt = "### Input: write a pandas dataframe script to drop null columns\n### Intent:"inputs = tokenizer(prompt, return_tensors="pt") # Run inferencewith torch.no_grad(): outputs = model.generate( inputs["input_ids"], max_new_tokens=6, temperature=0.1, do_sample=False, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id ) result = tokenizer.decode(outputs[0], skip_special_tokens=True)print(result)Developed by the 15m AI open-source organization as part of the high-performance micro-model initiative.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
text
http://www.apache.org/licenses/LICENSE-2.0Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
import torchfrom transformers import GPT2LMHeadModel, PreTrainedTokenizerFast # Load model and tokenizertokenizer = PreTrainedTokenizerFast.from_pretrained("15MAI/15m-pico-router")model = GPT2LMHeadModel.from_pretrained("15MAI/15m-pico-router")model.eval() # Prepare structured inputprompt = "### Input: write a pandas dataframe script to drop null columns\n### Intent:"inputs = tokenizer(prompt, return_tensors="pt") # Run inferencewith torch.no_grad(): outputs = model.generate( inputs["input_ids"], max_new_tokens=6, temperature=0.1, do_sample=False, pad_token_id=tokenizer.pad_token_id, eos_token_id=tokenizer.eos_token_id ) result = tokenizer.decode(outputs[0], skip_special_tokens=True)print(result)