Instructions to use shreyaschavan11/Akshar with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use shreyaschavan11/Akshar with PEFT:
```
Task type is invalid.
```

How to use shreyaschavan11/Akshar with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="shreyaschavan11/Akshar")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("shreyaschavan11/Akshar")
model = AutoModelForMultimodalLM.from_pretrained("shreyaschavan11/Akshar")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use shreyaschavan11/Akshar with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shreyaschavan11/Akshar"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shreyaschavan11/Akshar",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/shreyaschavan11/Akshar

SGLang

How to use shreyaschavan11/Akshar with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "shreyaschavan11/Akshar" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shreyaschavan11/Akshar",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "shreyaschavan11/Akshar" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shreyaschavan11/Akshar",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use shreyaschavan11/Akshar with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for shreyaschavan11/Akshar to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for shreyaschavan11/Akshar to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for shreyaschavan11/Akshar to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="shreyaschavan11/Akshar",
    max_seq_length=2048,
)

Docker Model Runner
How to use shreyaschavan11/Akshar with Docker Model Runner:
```
docker model run hf.co/shreyaschavan11/Akshar
```

Akshar / config.json

shreyaschavan11

Upload merged 4-bit Akshar model

168a25e 3 months ago

Raw

History Blame Contribute Delete

29.4 kB

	{
	"architectures": [
	"Gemma4ForConditionalGeneration"
	],
	"audio_config": {
	"_name_or_path": "",
	"architectures": null,
	"attention_chunk_size": 12,
	"attention_context_left": 13,
	"attention_context_right": 0,
	"attention_invalid_logits_value": -1000000000.0,
	"attention_logit_cap": 50.0,
	"chunk_size_feed_forward": 0,
	"conv_kernel_size": 5,
	"dtype": "float16",
	"gradient_clipping": 10000000000.0,
	"hidden_act": "silu",
	"hidden_size": 1024,
	"id2label": {
	"0": "LABEL_0",
	"1": "LABEL_1"
	},
	"initializer_range": 0.02,
	"is_encoder_decoder": false,
	"label2id": {
	"LABEL_0": 0,
	"LABEL_1": 1
	},
	"model_type": "gemma4_audio",
	"num_attention_heads": 8,
	"num_hidden_layers": 12,
	"output_attentions": false,
	"output_hidden_states": false,
	"output_proj_dims": 1536,
	"problem_type": null,
	"residual_weight": 0.5,
	"return_dict": true,
	"rms_norm_eps": 1e-06,
	"subsampling_conv_channels": [
	128,
	32
	],
	"use_clipped_linears": true
	},
	"audio_token_id": 258881,
	"boa_token_id": 256000,
	"boi_token_id": 255999,
	"dtype": "float16",
	"eoa_token_id": 258883,
	"eoa_token_index": 258883,
	"eoi_token_id": 258882,
	"eos_token_id": 1,
	"image_token_id": 258880,
	"initializer_range": 0.02,
	"model_name": "unsloth/gemma-4-E2B-it-unsloth-bnb-4bit",
	"model_type": "gemma4",
	"pad_token_id": 0,
	"quantization_config": {
	"bnb_4bit_compute_dtype": "float16",
	"bnb_4bit_quant_type": "nf4",
	"bnb_4bit_use_double_quant": true,
	"llm_int8_enable_fp32_cpu_offload": false,
	"llm_int8_has_fp16_weight": false,
	"llm_int8_skip_modules": [
	"model.language_model.layers.0.self_attn.k_proj",
	"model.language_model.layers.0.self_attn.q_proj",
	"model.language_model.layers.0.self_attn.v_proj",
	"model.language_model.layers.0.self_attn.o_proj",
	"model.language_model.layers.0.mlp.gate_proj",
	"model.language_model.layers.0.mlp.up_proj",
	"model.language_model.layers.0.mlp.down_proj",
	"model.language_model.layers.0.per_layer_input_gate",
	"model.language_model.layers.0.per_layer_projection",
	"model.language_model.layers.1.per_layer_input_gate",
	"model.language_model.layers.1.per_layer_projection",
	"model.language_model.layers.2.self_attn.k_proj",
	"model.language_model.layers.2.self_attn.q_proj",
	"model.language_model.layers.2.self_attn.v_proj",
	"model.language_model.layers.2.self_attn.o_proj",
	"model.language_model.layers.2.per_layer_input_gate",
	"model.language_model.layers.2.per_layer_projection",
	"model.language_model.layers.3.per_layer_input_gate",
	"model.language_model.layers.3.per_layer_projection",
	"model.language_model.layers.4.per_layer_input_gate",
	"model.language_model.layers.4.per_layer_projection",
	"model.language_model.layers.5.per_layer_input_gate",
	"model.language_model.layers.5.per_layer_projection",
	"model.language_model.layers.6.mlp.gate_proj",
	"model.language_model.layers.6.mlp.up_proj",
	"model.language_model.layers.6.mlp.down_proj",
	"model.language_model.layers.6.per_layer_input_gate",
	"model.language_model.layers.6.per_layer_projection",
	"model.language_model.layers.7.mlp.gate_proj",
	"model.language_model.layers.7.mlp.up_proj",
	"model.language_model.layers.7.mlp.down_proj",
	"model.language_model.layers.7.per_layer_input_gate",
	"model.language_model.layers.7.per_layer_projection",
	"model.language_model.layers.8.self_attn.k_proj",
	"model.language_model.layers.8.self_attn.q_proj",
	"model.language_model.layers.8.self_attn.v_proj",
	"model.language_model.layers.8.self_attn.o_proj",
	"model.language_model.layers.8.mlp.gate_proj",
	"model.language_model.layers.8.mlp.up_proj",
	"model.language_model.layers.8.mlp.down_proj",
	"model.language_model.layers.8.per_layer_input_gate",
	"model.language_model.layers.8.per_layer_projection",
	"model.language_model.layers.9.self_attn.k_proj",
	"model.language_model.layers.9.self_attn.q_proj",
	"model.language_model.layers.9.self_attn.v_proj",
	"model.language_model.layers.9.self_attn.o_proj",
	"model.language_model.layers.9.mlp.gate_proj",
	"model.language_model.layers.9.mlp.up_proj",
	"model.language_model.layers.9.mlp.down_proj",
	"model.language_model.layers.9.per_layer_input_gate",
	"model.language_model.layers.9.per_layer_projection",
	"model.language_model.layers.10.self_attn.k_proj",
	"model.language_model.layers.10.self_attn.q_proj",
	"model.language_model.layers.10.self_attn.v_proj",
	"model.language_model.layers.10.self_attn.o_proj",
	"model.language_model.layers.10.mlp.gate_proj",
	"model.language_model.layers.10.mlp.up_proj",
	"model.language_model.layers.10.mlp.down_proj",
	"model.language_model.layers.10.per_layer_input_gate",
	"model.language_model.layers.10.per_layer_projection",
	"model.language_model.layers.11.mlp.gate_proj",
	"model.language_model.layers.11.mlp.up_proj",
	"model.language_model.layers.11.mlp.down_proj",
	"model.language_model.layers.11.per_layer_input_gate",
	"model.language_model.layers.11.per_layer_projection",
	"model.language_model.layers.12.self_attn.k_proj",
	"model.language_model.layers.12.self_attn.q_proj",
	"model.language_model.layers.12.self_attn.v_proj",
	"model.language_model.layers.12.self_attn.o_proj",
	"model.language_model.layers.12.mlp.gate_proj",
	"model.language_model.layers.12.mlp.up_proj",
	"model.language_model.layers.12.mlp.down_proj",
	"model.language_model.layers.12.per_layer_input_gate",
	"model.language_model.layers.12.per_layer_projection",
	"model.language_model.layers.13.self_attn.k_proj",
	"model.language_model.layers.13.self_attn.q_proj",
	"model.language_model.layers.13.self_attn.v_proj",
	"model.language_model.layers.13.self_attn.o_proj",
	"model.language_model.layers.13.mlp.gate_proj",
	"model.language_model.layers.13.mlp.up_proj",
	"model.language_model.layers.13.mlp.down_proj",
	"model.language_model.layers.13.per_layer_input_gate",
	"model.language_model.layers.13.per_layer_projection",
	"model.language_model.layers.14.self_attn.k_proj",
	"model.language_model.layers.14.self_attn.q_proj",
	"model.language_model.layers.14.self_attn.v_proj",
	"model.language_model.layers.14.self_attn.o_proj",
	"model.language_model.layers.14.per_layer_input_gate",
	"model.language_model.layers.14.per_layer_projection",
	"model.language_model.layers.15.per_layer_input_gate",
	"model.language_model.layers.15.per_layer_projection",
	"model.language_model.layers.16.per_layer_input_gate",
	"model.language_model.layers.16.per_layer_projection",
	"model.language_model.layers.17.per_layer_input_gate",
	"model.language_model.layers.17.per_layer_projection",
	"model.language_model.layers.18.per_layer_input_gate",
	"model.language_model.layers.18.per_layer_projection",
	"model.language_model.layers.19.per_layer_input_gate",
	"model.language_model.layers.19.per_layer_projection",
	"model.language_model.layers.20.per_layer_input_gate",
	"model.language_model.layers.20.per_layer_projection",
	"model.language_model.layers.21.per_layer_input_gate",
	"model.language_model.layers.21.per_layer_projection",
	"model.language_model.layers.22.per_layer_input_gate",
	"model.language_model.layers.22.per_layer_projection",
	"model.language_model.layers.23.per_layer_input_gate",
	"model.language_model.layers.23.per_layer_projection",
	"model.language_model.layers.24.per_layer_input_gate",
	"model.language_model.layers.24.per_layer_projection",
	"model.language_model.layers.25.per_layer_input_gate",
	"model.language_model.layers.25.per_layer_projection",
	"model.language_model.layers.26.per_layer_input_gate",
	"model.language_model.layers.26.per_layer_projection",
	"model.language_model.layers.27.per_layer_input_gate",
	"model.language_model.layers.27.per_layer_projection",
	"model.language_model.layers.28.per_layer_input_gate",
	"model.language_model.layers.28.per_layer_projection",
	"model.language_model.layers.29.per_layer_input_gate",
	"model.language_model.layers.29.per_layer_projection",
	"model.language_model.layers.30.per_layer_input_gate",
	"model.language_model.layers.30.per_layer_projection",
	"model.language_model.layers.31.per_layer_input_gate",
	"model.language_model.layers.31.per_layer_projection",
	"model.language_model.layers.32.per_layer_input_gate",
	"model.language_model.layers.32.per_layer_projection",
	"model.language_model.layers.33.per_layer_input_gate",
	"model.language_model.layers.33.per_layer_projection",
	"model.language_model.layers.34.per_layer_input_gate",
	"model.language_model.layers.34.per_layer_projection",
	"model.language_model.per_layer_model_projection",
	"model.vision_tower.patch_embedder.input_proj",
	"model.vision_tower.encoder.layers.0.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.0.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.0.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.0.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.0.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.0.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.0.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.1.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.1.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.1.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.1.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.1.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.1.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.1.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.2.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.2.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.2.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.2.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.2.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.2.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.2.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.3.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.3.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.3.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.3.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.3.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.3.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.3.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.4.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.4.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.4.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.4.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.4.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.4.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.4.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.5.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.5.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.5.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.5.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.5.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.5.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.5.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.6.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.6.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.6.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.6.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.6.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.6.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.6.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.7.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.7.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.7.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.7.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.7.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.7.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.7.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.8.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.8.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.8.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.8.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.8.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.8.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.8.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.9.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.9.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.9.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.9.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.9.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.9.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.9.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.10.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.10.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.10.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.10.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.10.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.10.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.10.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.11.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.11.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.11.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.11.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.11.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.11.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.11.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.12.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.12.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.12.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.12.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.12.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.12.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.12.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.13.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.13.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.13.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.13.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.13.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.13.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.13.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.14.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.14.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.14.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.14.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.14.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.14.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.14.mlp.down_proj.linear",
	"model.vision_tower.encoder.layers.15.self_attn.q_proj.linear",
	"model.vision_tower.encoder.layers.15.self_attn.k_proj.linear",
	"model.vision_tower.encoder.layers.15.self_attn.v_proj.linear",
	"model.vision_tower.encoder.layers.15.self_attn.o_proj.linear",
	"model.vision_tower.encoder.layers.15.mlp.gate_proj.linear",
	"model.vision_tower.encoder.layers.15.mlp.up_proj.linear",
	"model.vision_tower.encoder.layers.15.mlp.down_proj.linear",
	"model.embed_vision.embedding_projection",
	"model.audio_tower.subsample_conv_projection.input_proj_linear",
	"model.audio_tower.layers.0.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.0.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.0.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.0.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.0.self_attn.q_proj.linear",
	"model.audio_tower.layers.0.self_attn.k_proj.linear",
	"model.audio_tower.layers.0.self_attn.v_proj.linear",
	"model.audio_tower.layers.0.self_attn.post.linear",
	"model.audio_tower.layers.0.self_attn.relative_k_proj",
	"model.audio_tower.layers.0.lconv1d.linear_start.linear",
	"model.audio_tower.layers.0.lconv1d.linear_end.linear",
	"model.audio_tower.layers.1.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.1.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.1.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.1.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.1.self_attn.q_proj.linear",
	"model.audio_tower.layers.1.self_attn.k_proj.linear",
	"model.audio_tower.layers.1.self_attn.v_proj.linear",
	"model.audio_tower.layers.1.self_attn.post.linear",
	"model.audio_tower.layers.1.self_attn.relative_k_proj",
	"model.audio_tower.layers.1.lconv1d.linear_start.linear",
	"model.audio_tower.layers.1.lconv1d.linear_end.linear",
	"model.audio_tower.layers.2.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.2.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.2.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.2.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.2.self_attn.q_proj.linear",
	"model.audio_tower.layers.2.self_attn.k_proj.linear",
	"model.audio_tower.layers.2.self_attn.v_proj.linear",
	"model.audio_tower.layers.2.self_attn.post.linear",
	"model.audio_tower.layers.2.self_attn.relative_k_proj",
	"model.audio_tower.layers.2.lconv1d.linear_start.linear",
	"model.audio_tower.layers.2.lconv1d.linear_end.linear",
	"model.audio_tower.layers.3.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.3.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.3.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.3.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.3.self_attn.q_proj.linear",
	"model.audio_tower.layers.3.self_attn.k_proj.linear",
	"model.audio_tower.layers.3.self_attn.v_proj.linear",
	"model.audio_tower.layers.3.self_attn.post.linear",
	"model.audio_tower.layers.3.self_attn.relative_k_proj",
	"model.audio_tower.layers.3.lconv1d.linear_start.linear",
	"model.audio_tower.layers.3.lconv1d.linear_end.linear",
	"model.audio_tower.layers.4.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.4.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.4.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.4.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.4.self_attn.q_proj.linear",
	"model.audio_tower.layers.4.self_attn.k_proj.linear",
	"model.audio_tower.layers.4.self_attn.v_proj.linear",
	"model.audio_tower.layers.4.self_attn.post.linear",
	"model.audio_tower.layers.4.self_attn.relative_k_proj",
	"model.audio_tower.layers.4.lconv1d.linear_start.linear",
	"model.audio_tower.layers.4.lconv1d.linear_end.linear",
	"model.audio_tower.layers.5.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.5.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.5.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.5.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.5.self_attn.q_proj.linear",
	"model.audio_tower.layers.5.self_attn.k_proj.linear",
	"model.audio_tower.layers.5.self_attn.v_proj.linear",
	"model.audio_tower.layers.5.self_attn.post.linear",
	"model.audio_tower.layers.5.self_attn.relative_k_proj",
	"model.audio_tower.layers.5.lconv1d.linear_start.linear",
	"model.audio_tower.layers.5.lconv1d.linear_end.linear",
	"model.audio_tower.layers.6.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.6.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.6.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.6.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.6.self_attn.q_proj.linear",
	"model.audio_tower.layers.6.self_attn.k_proj.linear",
	"model.audio_tower.layers.6.self_attn.v_proj.linear",
	"model.audio_tower.layers.6.self_attn.post.linear",
	"model.audio_tower.layers.6.self_attn.relative_k_proj",
	"model.audio_tower.layers.6.lconv1d.linear_start.linear",
	"model.audio_tower.layers.6.lconv1d.linear_end.linear",
	"model.audio_tower.layers.7.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.7.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.7.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.7.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.7.self_attn.q_proj.linear",
	"model.audio_tower.layers.7.self_attn.k_proj.linear",
	"model.audio_tower.layers.7.self_attn.v_proj.linear",
	"model.audio_tower.layers.7.self_attn.post.linear",
	"model.audio_tower.layers.7.self_attn.relative_k_proj",
	"model.audio_tower.layers.7.lconv1d.linear_start.linear",
	"model.audio_tower.layers.7.lconv1d.linear_end.linear",
	"model.audio_tower.layers.8.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.8.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.8.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.8.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.8.self_attn.q_proj.linear",
	"model.audio_tower.layers.8.self_attn.k_proj.linear",
	"model.audio_tower.layers.8.self_attn.v_proj.linear",
	"model.audio_tower.layers.8.self_attn.post.linear",
	"model.audio_tower.layers.8.self_attn.relative_k_proj",
	"model.audio_tower.layers.8.lconv1d.linear_start.linear",
	"model.audio_tower.layers.8.lconv1d.linear_end.linear",
	"model.audio_tower.layers.9.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.9.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.9.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.9.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.9.self_attn.q_proj.linear",
	"model.audio_tower.layers.9.self_attn.k_proj.linear",
	"model.audio_tower.layers.9.self_attn.v_proj.linear",
	"model.audio_tower.layers.9.self_attn.post.linear",
	"model.audio_tower.layers.9.self_attn.relative_k_proj",
	"model.audio_tower.layers.9.lconv1d.linear_start.linear",
	"model.audio_tower.layers.9.lconv1d.linear_end.linear",
	"model.audio_tower.layers.10.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.10.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.10.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.10.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.10.self_attn.q_proj.linear",
	"model.audio_tower.layers.10.self_attn.k_proj.linear",
	"model.audio_tower.layers.10.self_attn.v_proj.linear",
	"model.audio_tower.layers.10.self_attn.post.linear",
	"model.audio_tower.layers.10.self_attn.relative_k_proj",
	"model.audio_tower.layers.10.lconv1d.linear_start.linear",
	"model.audio_tower.layers.10.lconv1d.linear_end.linear",
	"model.audio_tower.layers.11.feed_forward1.ffw_layer_1.linear",
	"model.audio_tower.layers.11.feed_forward1.ffw_layer_2.linear",
	"model.audio_tower.layers.11.feed_forward2.ffw_layer_1.linear",
	"model.audio_tower.layers.11.feed_forward2.ffw_layer_2.linear",
	"model.audio_tower.layers.11.self_attn.q_proj.linear",
	"model.audio_tower.layers.11.self_attn.k_proj.linear",
	"model.audio_tower.layers.11.self_attn.v_proj.linear",
	"model.audio_tower.layers.11.self_attn.post.linear",
	"model.audio_tower.layers.11.self_attn.relative_k_proj",
	"model.audio_tower.layers.11.lconv1d.linear_start.linear",
	"model.audio_tower.layers.11.lconv1d.linear_end.linear",
	"model.audio_tower.output_proj",
	"model.embed_audio.embedding_projection",
	"lm_head"
	],
	"llm_int8_threshold": 6.0,
	"load_in_4bit": true,
	"load_in_8bit": false,
	"quant_method": "bitsandbytes"
	},
	"text_config": {
	"attention_bias": false,
	"attention_dropout": 0.0,
	"attention_k_eq_v": false,
	"bos_token_id": 2,
	"dtype": "float16",
	"enable_moe_block": false,
	"eos_token_id": 1,
	"expert_intermediate_size": null,
	"final_logit_softcapping": 30.0,
	"global_head_dim": 512,
	"head_dim": 256,
	"hidden_activation": "gelu_pytorch_tanh",
	"hidden_size": 1536,
	"hidden_size_per_layer_input": 256,
	"initializer_range": 0.02,
	"intermediate_size": 6144,
	"layer_types": [
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention"
	],
	"max_position_embeddings": 131072,
	"model_type": "gemma4_text",
	"moe_intermediate_size": null,
	"num_attention_heads": 8,
	"num_experts": null,
	"num_global_key_value_heads": null,
	"num_hidden_layers": 35,
	"num_key_value_heads": 1,
	"num_kv_shared_layers": 20,
	"pad_token_id": 0,
	"rms_norm_eps": 1e-06,
	"rope_parameters": {
	"full_attention": {
	"partial_rotary_factor": 0.25,
	"rope_theta": 1000000.0,
	"rope_type": "proportional"
	},
	"sliding_attention": {
	"rope_theta": 10000.0,
	"rope_type": "default"
	}
	},
	"sliding_window": 512,
	"tie_word_embeddings": true,
	"top_k_experts": null,
	"use_bidirectional_attention": null,
	"use_cache": true,
	"use_double_wide_mlp": true,
	"vocab_size": 262144,
	"vocab_size_per_layer_input": 262144
	},
	"tie_word_embeddings": true,
	"transformers_version": "5.5.0",
	"unsloth_fixed": true,
	"unsloth_version": "2026.4.4",
	"use_cache": true,
	"video_token_id": 258884,
	"vision_config": {
	"_name_or_path": "",
	"architectures": null,
	"attention_bias": false,
	"attention_dropout": 0.0,
	"chunk_size_feed_forward": 0,
	"default_output_length": 280,
	"dtype": "float16",
	"global_head_dim": 64,
	"head_dim": 64,
	"hidden_activation": "gelu_pytorch_tanh",
	"hidden_size": 768,
	"id2label": {
	"0": "LABEL_0",
	"1": "LABEL_1"
	},
	"initializer_range": 0.02,
	"intermediate_size": 3072,
	"is_encoder_decoder": false,
	"label2id": {
	"LABEL_0": 0,
	"LABEL_1": 1
	},
	"max_position_embeddings": 131072,
	"model_type": "gemma4_vision",
	"num_attention_heads": 12,
	"num_hidden_layers": 16,
	"num_key_value_heads": 12,
	"output_attentions": false,
	"output_hidden_states": false,
	"patch_size": 16,
	"pooling_kernel_size": 3,
	"position_embedding_size": 10240,
	"problem_type": null,
	"return_dict": true,
	"rms_norm_eps": 1e-06,
	"rope_parameters": {
	"rope_theta": 100.0,
	"rope_type": "default"
	},
	"standardize": false,
	"use_clipped_linears": true
	},
	"vision_soft_tokens_per_image": 280
	}