Instructions to use EchoLabs33/zamba2-7b-instruct-hxq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EchoLabs33/zamba2-7b-instruct-hxq with Transformers:

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

pipe = pipeline("text-generation", model="EchoLabs33/zamba2-7b-instruct-hxq")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("EchoLabs33/zamba2-7b-instruct-hxq")
model = AutoModelForCausalLM.from_pretrained("EchoLabs33/zamba2-7b-instruct-hxq", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use EchoLabs33/zamba2-7b-instruct-hxq with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EchoLabs33/zamba2-7b-instruct-hxq"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EchoLabs33/zamba2-7b-instruct-hxq",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/EchoLabs33/zamba2-7b-instruct-hxq

SGLang

How to use EchoLabs33/zamba2-7b-instruct-hxq 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 "EchoLabs33/zamba2-7b-instruct-hxq" \
    --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": "EchoLabs33/zamba2-7b-instruct-hxq",
		"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 "EchoLabs33/zamba2-7b-instruct-hxq" \
        --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": "EchoLabs33/zamba2-7b-instruct-hxq",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use EchoLabs33/zamba2-7b-instruct-hxq with Docker Model Runner:
```
docker model run hf.co/EchoLabs33/zamba2-7b-instruct-hxq
```

zamba2-7b-instruct-hxq / config.json

voidstream

Upload config.json with huggingface_hub

f57f5f1 verified 4 months ago

Raw

History Blame Contribute Delete

13 kB

	{
	"architectures": [
	"Zamba2ForCausalLM"
	],
	"adapter_rank": 128,
	"add_bias_linear": false,
	"attention_dropout": 0.0,
	"attention_head_dim": 224,
	"attention_hidden_size": 7168,
	"bos_token_id": 1,
	"chunk_size": 256,
	"eos_token_id": 2,
	"ffn_hidden_size": 14336,
	"hidden_act": "gelu",
	"hidden_size": 3584,
	"hybrid_layer_ids": [
	6,
	11,
	17,
	23,
	29,
	35,
	41,
	47,
	53,
	59,
	65,
	71,
	77
	],
	"initializer_range": 0.02,
	"intermediate_size": 14336,
	"kv_channels": 112,
	"layers_block_type": [
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba"
	],
	"mamba_d_conv": 4,
	"mamba_d_state": 64,
	"mamba_expand": 2,
	"mamba_headdim": 64,
	"mamba_ngroups": 2,
	"max_position_embeddings": 4096,
	"model_type": "zamba2",
	"n_mamba_heads": 112,
	"num_attention_heads": 32,
	"num_hidden_layers": 81,
	"num_key_value_heads": 32,
	"num_logits_to_keep": 1,
	"num_mem_blocks": 2,
	"num_query_groups": 32,
	"pad_token_id": 0,
	"rms_norm_eps": 1e-05,
	"rope_theta": 10000,
	"time_step_floor": 0.0001,
	"time_step_limit": null,
	"time_step_max": 0.1,
	"time_step_min": 0.001,
	"transformers_version": "4.49.0.dev0",
	"use_cache": true,
	"use_conv_bias": true,
	"use_long_context": false,
	"use_mem_rope": true,
	"use_shared_attention_adapter": false,
	"use_shared_mlp_adapter": true,
	"vocab_size": 32000,
	"quantization_config": {
	"quant_method": "hxq",
	"codebook_size": 256,
	"sidecar_enabled": true,
	"exact_patterns": [
	"embed_tokens",
	"embed_positions",
	"wte",
	"wpe",
	"lm_head",
	"layernorm",
	"layer_norm",
	"norm",
	"backbone.embedding"
	],
	"compressed_modules": [
	"model.layers.0.mamba.in_proj",
	"model.layers.0.mamba.out_proj",
	"model.layers.10.mamba.in_proj",
	"model.layers.10.mamba.out_proj",
	"model.layers.11.linear",
	"model.layers.11.mamba_decoder.mamba.in_proj",
	"model.layers.11.mamba_decoder.mamba.out_proj",
	"model.layers.11.shared_transformer.feed_forward.down_proj",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.11.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.11.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.1.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.1.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.3.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.3.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.5.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.5.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.7.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.7.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.9.0",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj_adapter_list.9.1",
	"model.layers.11.shared_transformer.feed_forward.gate_up_proj",
	"model.layers.11.shared_transformer.self_attn.k_proj",
	"model.layers.11.shared_transformer.self_attn.o_proj",
	"model.layers.11.shared_transformer.self_attn.q_proj",
	"model.layers.11.shared_transformer.self_attn.v_proj",
	"model.layers.12.mamba.in_proj",
	"model.layers.12.mamba.out_proj",
	"model.layers.13.mamba.in_proj",
	"model.layers.13.mamba.out_proj",
	"model.layers.14.mamba.in_proj",
	"model.layers.14.mamba.out_proj",
	"model.layers.15.mamba.in_proj",
	"model.layers.15.mamba.out_proj",
	"model.layers.16.mamba.in_proj",
	"model.layers.16.mamba.out_proj",
	"model.layers.17.linear",
	"model.layers.17.mamba_decoder.mamba.in_proj",
	"model.layers.17.mamba_decoder.mamba.out_proj",
	"model.layers.18.mamba.in_proj",
	"model.layers.18.mamba.out_proj",
	"model.layers.19.mamba.in_proj",
	"model.layers.19.mamba.out_proj",
	"model.layers.1.mamba.in_proj",
	"model.layers.1.mamba.out_proj",
	"model.layers.20.mamba.in_proj",
	"model.layers.20.mamba.out_proj",
	"model.layers.21.mamba.in_proj",
	"model.layers.21.mamba.out_proj",
	"model.layers.22.mamba.in_proj",
	"model.layers.22.mamba.out_proj",
	"model.layers.23.linear",
	"model.layers.23.mamba_decoder.mamba.in_proj",
	"model.layers.23.mamba_decoder.mamba.out_proj",
	"model.layers.24.mamba.in_proj",
	"model.layers.24.mamba.out_proj",
	"model.layers.25.mamba.in_proj",
	"model.layers.25.mamba.out_proj",
	"model.layers.26.mamba.in_proj",
	"model.layers.26.mamba.out_proj",
	"model.layers.27.mamba.in_proj",
	"model.layers.27.mamba.out_proj",
	"model.layers.28.mamba.in_proj",
	"model.layers.28.mamba.out_proj",
	"model.layers.29.linear",
	"model.layers.29.mamba_decoder.mamba.in_proj",
	"model.layers.29.mamba_decoder.mamba.out_proj",
	"model.layers.2.mamba.in_proj",
	"model.layers.2.mamba.out_proj",
	"model.layers.30.mamba.in_proj",
	"model.layers.30.mamba.out_proj",
	"model.layers.31.mamba.in_proj",
	"model.layers.31.mamba.out_proj",
	"model.layers.32.mamba.in_proj",
	"model.layers.32.mamba.out_proj",
	"model.layers.33.mamba.in_proj",
	"model.layers.33.mamba.out_proj",
	"model.layers.34.mamba.in_proj",
	"model.layers.34.mamba.out_proj",
	"model.layers.35.linear",
	"model.layers.35.mamba_decoder.mamba.in_proj",
	"model.layers.35.mamba_decoder.mamba.out_proj",
	"model.layers.36.mamba.in_proj",
	"model.layers.36.mamba.out_proj",
	"model.layers.37.mamba.in_proj",
	"model.layers.37.mamba.out_proj",
	"model.layers.38.mamba.in_proj",
	"model.layers.38.mamba.out_proj",
	"model.layers.39.mamba.in_proj",
	"model.layers.39.mamba.out_proj",
	"model.layers.3.mamba.in_proj",
	"model.layers.3.mamba.out_proj",
	"model.layers.40.mamba.in_proj",
	"model.layers.40.mamba.out_proj",
	"model.layers.41.linear",
	"model.layers.41.mamba_decoder.mamba.in_proj",
	"model.layers.41.mamba_decoder.mamba.out_proj",
	"model.layers.42.mamba.in_proj",
	"model.layers.42.mamba.out_proj",
	"model.layers.43.mamba.in_proj",
	"model.layers.43.mamba.out_proj",
	"model.layers.44.mamba.in_proj",
	"model.layers.44.mamba.out_proj",
	"model.layers.45.mamba.in_proj",
	"model.layers.45.mamba.out_proj",
	"model.layers.46.mamba.in_proj",
	"model.layers.46.mamba.out_proj",
	"model.layers.47.linear",
	"model.layers.47.mamba_decoder.mamba.in_proj",
	"model.layers.47.mamba_decoder.mamba.out_proj",
	"model.layers.48.mamba.in_proj",
	"model.layers.48.mamba.out_proj",
	"model.layers.49.mamba.in_proj",
	"model.layers.49.mamba.out_proj",
	"model.layers.4.mamba.in_proj",
	"model.layers.4.mamba.out_proj",
	"model.layers.50.mamba.in_proj",
	"model.layers.50.mamba.out_proj",
	"model.layers.51.mamba.in_proj",
	"model.layers.51.mamba.out_proj",
	"model.layers.52.mamba.in_proj",
	"model.layers.52.mamba.out_proj",
	"model.layers.53.linear",
	"model.layers.53.mamba_decoder.mamba.in_proj",
	"model.layers.53.mamba_decoder.mamba.out_proj",
	"model.layers.54.mamba.in_proj",
	"model.layers.54.mamba.out_proj",
	"model.layers.55.mamba.in_proj",
	"model.layers.55.mamba.out_proj",
	"model.layers.56.mamba.in_proj",
	"model.layers.56.mamba.out_proj",
	"model.layers.57.mamba.in_proj",
	"model.layers.57.mamba.out_proj",
	"model.layers.58.mamba.in_proj",
	"model.layers.58.mamba.out_proj",
	"model.layers.59.linear",
	"model.layers.59.mamba_decoder.mamba.in_proj",
	"model.layers.59.mamba_decoder.mamba.out_proj",
	"model.layers.5.mamba.in_proj",
	"model.layers.5.mamba.out_proj",
	"model.layers.60.mamba.in_proj",
	"model.layers.60.mamba.out_proj",
	"model.layers.61.mamba.in_proj",
	"model.layers.61.mamba.out_proj",
	"model.layers.62.mamba.in_proj",
	"model.layers.62.mamba.out_proj",
	"model.layers.63.mamba.in_proj",
	"model.layers.63.mamba.out_proj",
	"model.layers.64.mamba.in_proj",
	"model.layers.64.mamba.out_proj",
	"model.layers.65.linear",
	"model.layers.65.mamba_decoder.mamba.in_proj",
	"model.layers.65.mamba_decoder.mamba.out_proj",
	"model.layers.66.mamba.in_proj",
	"model.layers.66.mamba.out_proj",
	"model.layers.67.mamba.in_proj",
	"model.layers.67.mamba.out_proj",
	"model.layers.68.mamba.in_proj",
	"model.layers.68.mamba.out_proj",
	"model.layers.69.mamba.in_proj",
	"model.layers.69.mamba.out_proj",
	"model.layers.6.linear",
	"model.layers.6.mamba_decoder.mamba.in_proj",
	"model.layers.6.mamba_decoder.mamba.out_proj",
	"model.layers.6.shared_transformer.feed_forward.down_proj",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.0.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.0.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.10.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.10.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.12.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.12.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.2.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.2.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.4.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.4.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.6.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.6.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.8.0",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj_adapter_list.8.1",
	"model.layers.6.shared_transformer.feed_forward.gate_up_proj",
	"model.layers.6.shared_transformer.self_attn.k_proj",
	"model.layers.6.shared_transformer.self_attn.o_proj",
	"model.layers.6.shared_transformer.self_attn.q_proj",
	"model.layers.6.shared_transformer.self_attn.v_proj",
	"model.layers.70.mamba.in_proj",
	"model.layers.70.mamba.out_proj",
	"model.layers.71.linear",
	"model.layers.71.mamba_decoder.mamba.in_proj",
	"model.layers.71.mamba_decoder.mamba.out_proj",
	"model.layers.72.mamba.in_proj",
	"model.layers.72.mamba.out_proj",
	"model.layers.73.mamba.in_proj",
	"model.layers.73.mamba.out_proj",
	"model.layers.74.mamba.in_proj",
	"model.layers.74.mamba.out_proj",
	"model.layers.75.mamba.in_proj",
	"model.layers.75.mamba.out_proj",
	"model.layers.76.mamba.in_proj",
	"model.layers.76.mamba.out_proj",
	"model.layers.77.linear",
	"model.layers.77.mamba_decoder.mamba.in_proj",
	"model.layers.77.mamba_decoder.mamba.out_proj",
	"model.layers.78.mamba.in_proj",
	"model.layers.78.mamba.out_proj",
	"model.layers.79.mamba.in_proj",
	"model.layers.79.mamba.out_proj",
	"model.layers.7.mamba.in_proj",
	"model.layers.7.mamba.out_proj",
	"model.layers.80.mamba.in_proj",
	"model.layers.80.mamba.out_proj",
	"model.layers.8.mamba.in_proj",
	"model.layers.8.mamba.out_proj",
	"model.layers.9.mamba.in_proj",
	"model.layers.9.mamba.out_proj"
	],
	"n_clusters": 4096,
	"vector_dim": 2,
	"index_packing": "12bit"
	}
	}