Instructions to use ggml-org-dev/tiny-Bespoke-Nimble-9B-v3-for-testing with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use ggml-org-dev/tiny-Bespoke-Nimble-9B-v3-for-testing with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("ggml-org-dev/tiny-Qwen3.5-9B-for-testing") model = PeftModel.from_pretrained(base_model, "ggml-org-dev/tiny-Bespoke-Nimble-9B-v3-for-testing") - Notebooks
- Google Colab
- Kaggle
tiny-Bespoke-Nimble-9B-v3-for-testing
This is a modified, damaged copy of bespokelabs/Bespoke-Nimble-9B-v3, made only to test software. Its output is gibberish. Do not use it for anything else.
Bespoke-Nimble-9B-v3 is a LoRA adapter for Qwen/Qwen3.5-9B. This copy is a LoRA adapter for the tiny base model ggml-org-dev/tiny-Qwen3.5-9B-for-testing, a sliced copy of Qwen3.5-9B. It keeps the schema config, prompt layout and LoRA layout of Bespoke-Nimble-9B-v3, so it runs the same code paths as the real model in llama.cpp (nimble decision type: the converter downloads the base model named in adapter_config.json and merges the adapter), but it is small enough for CI.
GGUF version: ggml-org-dev/tiny-Bespoke-Nimble-9B-v3-for-testing-GGUF
Changes from the original adapter
Sliced with slice_qwen35_lora.py, which follows the slice-model skill of llama.cpp:
base_model_name_or_pathpoints at ggml-org-dev/tiny-Qwen3.5-9B-for-testing- LoRA tensors of 4 of 32 layers (0, 1, 2, 3), narrowed with the same indices as the base model: A keeps the input columns of its base weight, B its output rows (rank 64 and alpha kept)
schema_config.json:modelandrevisionpoint at the sliced base
License
Derived from Bespoke-Nimble-9B-v3 by Bespoke Labs, under CC BY-NC 4.0 (no commercial use). The base model, Qwen3.5-9B, is licensed separately under Apache 2.0.
- Downloads last month
- -
Model tree for ggml-org-dev/tiny-Bespoke-Nimble-9B-v3-for-testing
Base model
Qwen/Qwen3.5-9B-Base