GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
CrowdMind
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
CrowdMind
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
Property Value
Model name Fred-9B Base model Qwen 3.5 9B Fine-tuning method LoRA LoRA rank 128 LoRA alpha 256 LoRA dropout 0.05 Batch size 4 Context length 2,048 tokens Learning rate 2e-4 Optimizer AdamW 8-bit Warmup steps 30 Maximum training steps 300 Training tokens ~3.21M Final training loss 0.0833 Final gradient norm 0.470 Training time ~1h 33m
Fred-9B was fine-tuned for 300 steps. The training run used a 2,048-token context length and a batch size of 4.
The learning-rate schedule used a short warmup followed by decay across the training run.
0.08337.41e-70.4703,214,912300 / 3001h 32m 45sNo evaluation dataset was configured for this training run, so an evaluation loss was not recorded.
One of the training checkpoints was tested with simple arithmetic and order-of-operations questions.
Prompt
text
Please solve this for me i need help: 2+1*3(1+1)+3
Fred-9B
text
The final answer is 11.
The model correctly handles the parentheses and multiplication before addition:
text
1 + 1 = 23 × 2 = 62 + 6 + 3 = 11
Prompt
text
Please solve this for me i need help: 3+1+1*4(1+1)+3*1
Fred-9B
text
The final answer is 15.
The model breaks the expression down using standard order of operations and reaches the correct result.
Fred-9B is intended for:
Because Fred-9B is an experimental fine-tune, its behavior may vary substantially depending on the prompt and task.
Fred-9B has several important limitations:
For serious evaluation, use a held-out evaluation dataset and established benchmarks rather than relying only on training loss.
For future versions of Fred, consider evaluating:
A separate validation/evaluation dataset should be configured for future training runs so that training loss can be compared against evaluation loss.
Fred-9B can be used with an inference stack that supports the model's exported format.
If the model is exported to GGUF, it can be used with compatible local
inference tools such as llama.cpp-based applications.
Example concept:
bash
# Example only — replace with the actual exported model filename./llama-cli -m Fred-9B.gguf
The exact command depends on the exported model format and inference framework.
For local deployment, quantized versions can significantly reduce memory requirements.
Possible variants include:
The best choice depends on available VRAM/RAM and the desired balance between speed, memory usage, and model quality.
The training run used approximately the following configuration:
yaml
model: Qwen 3.5 9Bmethod: LoRAtraining:batch_size: 4learning_rate: 0.0002optimizer: AdamW 8-bitcontext_length: 2048warmup_steps: 30max_steps: 300lora:rank: 128alpha: 256dropout: 0.05variant: lora
Fred-9B is based on Qwen 3.5 9B.
All original model licensing, attribution, usage restrictions, and terms from the Qwen base model remain important when using or redistributing Fred-9B. Check the official Qwen model license and documentation before publishing or deploying the model.
Experimental --- first training run
Fred-9B is currently an experimental fine-tune. Future versions can improve the dataset, evaluation process, training duration, instruction formatting, and inference quality.
Fred-9B --- a personal 9B fine-tune built on Qwen 3.5 9B.