GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
wisdompan
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
wisdompan
Model Tree
Qwen/Qwen3.5-2B
Input Modalities
Output Modalities
Supported Functionality
| Item | Value |
|---|---|
| Base model | Qwen/Qwen3.5-2B |
| Parameters | 1,881,825,088 |
| Fine-tuning | Full parameter supervised fine-tuning |
| Domains | Math, code, instruction following |
| Precision | bfloat16 |
| Context length used for SFT | 1,024 tokens |
| Checkpoint | 1 epoch / step 375 |
| Weight format | SafeTensors |
The checkpoint was trained with Adafactor, a cosine learning-rate schedule,
peak learning rate 1e-5, weight decay 0.1, gradient accumulation 8,
gradient checkpointing, and seed 20260902. The final recorded training loss
was 0.6251 and token accuracy was 0.8485.
This model is not presented as a production-ready assistant or as the final result of the MOPD experiment.
Qwen3.5 requires a recent Transformers build. The serving flow follows the upstream Qwen3.5 recommendation:
bash
pip install "transformers[serving] @ git+https://github.com/huggingface/transformers.git@main"transformers serve \--force-model wisdompan/qwen35-2b-mopd-mixsft \--port 8000 \--continuous-batching
Then call the OpenAI-compatible endpoint:
python
from openai import OpenAIclient = OpenAI(base_url="http://localhost:8000/v1", api_key="unused")response = client.chat.completions.create(model="wisdompan/qwen35-2b-mopd-mixsft",messages=[{"role": "user", "content": "Solve: 2x + 3 = 11"}],)print(response.choices[0].message.content)
Only a six-example pipeline sanity check was bundled with this checkpoint. It is useful for confirming that generation and reward routing work, but it is too small to report as a benchmark. Broader evaluation will accompany the final teacher and student releases.
text
Qwen/Qwen3.5-2B-> MixSFT (this repository)-> Math / Code / IF teacher RL-> MOPD student distillation
This derivative is released under the Apache License 2.0, following the base model license. Users must also comply with the terms and usage requirements of the upstream Qwen3.5-2B model.