Downloads · 30 days
9
30% of all-time downloads
LangJiaqi77/Voicemem-Qwen3_6-35B-A3B-QLoRA-v2
Voicemem-Qwen3_6-35B-A3B-QLoRA-v2 is a text generation model from LangJiaqi77. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This directory is the release manifest for the adapter-only checkpoint at training step 3318. The model weights are published separately on Hugging Face; they are intentionally not committed to this code repository.
Downloads · 30 days
9
30% of all-time downloads
All-time downloads
30
Public
Repo size
180 MB
Likes
0
Public
Click a slice to open those files.
.safetensors180 MB · 100%
From the Hugging Face model README
This directory is the release manifest for the adapter-only checkpoint at training step 3318. The model weights are published separately on Hugging Face; they are intentionally not committed to this code repository.
https://huggingface.co/<organisation>/voicemem-qwen3.6-35b-a3b-qlora-v2 before releasecheckpoint-3318, 2 epochs, 3318 stepsInstall a compatible transformers, peft, and the dependencies required by the upstream base model. Download the adapter from Hugging Face, then load it onto the permitted base model:
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base_model_id = "<official-base-model-hub-id>"
adapter_id = "<organisation>/voicemem-qwen3.6-35b-a3b-qlora-v2"
tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True)
base = AutoModelForCausalLM.from_pretrained(base_model_id, trust_remote_code=True)
model = PeftModel.from_pretrained(base, adapter_id)
Do not treat this adapter as a standalone model. Users must obtain the base model under its own license and access conditions.
The Hugging Face model repository should contain only:
adapter_model.safetensorsadapter_config.json with a public base-model identifierOptimizer shards, DeepSpeed state, RNG state, scheduler state, and training logs are not model artifacts and must not be uploaded as model weights.
On AudioMC INFERENCE_MEMORY (132 conversations; 233 rubric criteria), checkpoint-3318 satisfies 97 / 233 (41.6%) criteria. Under the same protocol, GPT-4o-mini satisfies 96 / 233 (41.2%). The adapter improves by 1 criterion, or +0.43 percentage points. GPT-4o-mini is used as the judge for both runs.
This is a research adapter. It has not been evaluated for safety-critical use, does not replace the base model's safety documentation, and may reproduce biases or errors from both training data and the base model.