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
```

voidstream commited on Apr 2

Commit

85b3b4e

verified ·

1 Parent(s): 1658dad

Upload zamba2_7b_comparison.json with huggingface_hub

Browse files

Files changed (1) hide show

zamba2_7b_comparison.json +31 -61

zamba2_7b_comparison.json CHANGED Viewed

@@ -1,71 +1,41 @@
 {
-  "experiment": "WO-BNB-BASELINE-01",
   "model": "Zamba2-7B-Instruct",
-  "eval_params": {
-    "dataset": "wikitext/wikitext-2-raw-v1/test",
-    "max_length": 2048,
-    "stride": 512,
-    "speed_iters": 100,
-    "speed_seq_len": 512
-  },
-  "configs": {
-    "dense_bf16": {
-      "name": "Dense BF16",
-      "ppl": 3.7216,
-      "n_tokens": 334660,
-      "n_chunks": 651,
       "vram_load_mb": 14032.0,
-      "vram_peak_eval_mb": 16173.8,
-      "vram_peak_speed_mb": 14396.5,
-      "tokens_per_sec": 4385.2,
-      "speed_test_elapsed_s": 11.68,
-      "config_elapsed_s": 297.3
     },
-    "bnb_8bit": {
-      "name": "bnb 8-bit (LLM.int8)",
-      "ppl": 3.7335,
-      "n_tokens": 334660,
-      "n_chunks": 651,
-      "vram_load_mb": 7831.3,
-      "vram_peak_eval_mb": 10011.6,
-      "vram_peak_speed_mb": 8188.0,
-      "tokens_per_sec": 2165.3,
-      "speed_test_elapsed_s": 23.65,
-      "config_elapsed_s": 334.3
     },
-    "bnb_4bit_nf4": {
-      "name": "bnb 4-bit NF4",
-      "ppl": 3.8916,
-      "n_tokens": 334660,
-      "n_chunks": 651,
-      "vram_load_mb": 5106.6,
-      "vram_peak_eval_mb": 7240.4,
-      "vram_peak_speed_mb": 5590.6,
-      "tokens_per_sec": 3468.4,
-      "speed_test_elapsed_s": 14.76,
-      "config_elapsed_s": 332.6
     },
-    "hxq_2dvq_k4096": {
-      "name": "HXQ 2D VQ k=4096",
-      "ppl": 3.8454,
-      "n_tokens": 334660,
-      "n_chunks": 651,
-      "vram_load_mb": 14040.1,
-      "vram_peak_eval_mb": 16173.8,
-      "vram_peak_speed_mb": 14396.5,
-      "tokens_per_sec": 4384.5,
-      "speed_test_elapsed_s": 11.68,
-      "config_elapsed_s": 343.0
     }
-  },
   "cost": {
-    "wall_time_s": 1311.822,
-    "cpu_time_s": 1440.214,
-    "peak_memory_mb": 16490.8,
-    "python_version": "3.10.12",
-    "hostname": "74efa4ce-5ef6-427c-b27b-336ce89875b5",
-    "timestamp_start": "2026-04-01T21:29:26",
-    "timestamp_end": "2026-04-01T21:51:18",
-    "gpu": "NVIDIA GeForce RTX 4090"
   }
 }

 {
+  "work_order": "WO-BUFFERED-FORWARD-01",
   "model": "Zamba2-7B-Instruct",
+  "gpu": "RTX 3090 24GB",
+  "results": [
+    {
+      "config": "Dense BF16",
+      "ppl": 4.8211,
+      "tok_s": 1445.9,
       "vram_load_mb": 14032.0,
+      "vram_peak_mb": 14686.0
     },
+    {
+      "config": "bnb 4-bit NF4",
+      "ppl": 5.0664,
+      "tok_s": 1579.4,
+      "vram_load_mb": 5129.4,
+      "vram_peak_mb": 5904.0
     },
+    {
+      "config": "HXQ 12-bit packed (buffered)",
+      "ppl": 5.0153,
+      "tok_s": 1763.7,
+      "vram_load_mb": 5657.2,
+      "vram_warmup_mb": 5864.6,
+      "vram_peak_mb": 6510.7
     },
+    {
+      "config": "HXQ 12-bit packed (fused)",
+      "ppl": 5.0228,
+      "tok_s": 283.5,
+      "vram_load_mb": 6079.5,
+      "vram_peak_mb": 7108.9
     }
+  ],
   "cost": {
+    "wall_time_s": -0.0,
+    "hostname": "8552b21e-f647-40e8-ab36-30dc6fb62665",
+    "timestamp": "2026-04-02T14:31:44"
   }
 }