csm-1b-MrDragonFoxElise-q4

model, processor = FastModel.from_pretrained(
    model_name = "unsloth/csm-1b",
    max_seq_length= 2048, # Choose any for long context!
    dtype = None, # Leave as None for auto-detection
    auto_model = CsmForConditionalGeneration,
    load_in_4bit = False, # Select True for 4bit - reduces memory usage
)

model = FastModel.get_peft_model(
    model,
    r = 32, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128
    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
                      "gate_proj", "up_proj", "down_proj",],
    lora_alpha = 32,
    lora_dropout = 0, # Supports any, but = 0 is optimized
    bias = "none",    # Supports any, but = "none" is optimized
    # [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes!
    use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context
    random_state = 3407,
    use_rslora = False,  # We support rank stabilized LoRA
    loftq_config = None, # And LoftQ
)

trainer = Trainer(
    model = model,
    train_dataset = processed_ds,
    args = TrainingArguments(
        per_device_train_batch_size = 2,
        gradient_accumulation_steps = 4,
        warmup_steps = 5,
        max_steps = 60,
        learning_rate = 2e-4,
        fp16 = not is_bfloat16_supported(),
        bf16 = is_bfloat16_supported(),
        logging_steps = 1,
        optim = "adamw_8bit",
        weight_decay = 0.001, # Turn this on if overfitting
        lr_scheduler_type = "linear",
        seed = 3407,
        output_dir = "outputs",
        report_to = "none", # Use TrackIO/WandB etc
    ),
)

Sample generation:

text = "Sesame is a super cool TTS model which can be fine tuned with Unsloth."

speaker_id = 0
# Another equivalent way to prepare the inputs
conversation = [
    {"role": str(speaker_id), "content": [{"type": "text", "text": text}]},
]
audio_values = model.generate(
    **processor.apply_chat_template(
        conversation,
        tokenize=True,
        return_dict=True,
    ).to("cuda"),
    max_new_tokens=125, # 125 tokens is 10 seconds of audio, for longer speech increase this
    # play with these parameters to tweak results
    # depth_decoder_top_k=0,
    # depth_decoder_top_p=0.9,
    # depth_decoder_do_sample=True,
    # depth_decoder_temperature=0.9,
    # top_k=0,
    # top_p=1.0,
    # temperature=0.9,
    # do_sample=True,
    #########################################################
    output_audio=True
)
audio = audio_values[0].to(torch.float32).cpu().numpy()
sf.write("example_without_context.wav", audio, 24000)
display(Audio(audio, rate=24000))

---

base_model: unsloth/csm-1b tags:

• text-generation-inference
• transformers
• unsloth
• csm license: apache-2.0 language:
• en

---

Uploaded finetuned model

• Developed by: xTronz
• License: apache-2.0
• Finetuned from model : unsloth/csm-1b

This csm model was trained 2x faster with Unsloth and Huggingface's TRL library.

<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>