Instructions to use ValiantLabs/Muse-Glimmer-30B-Esper4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ValiantLabs/Muse-Glimmer-30B-Esper4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="ValiantLabs/Muse-Glimmer-30B-Esper4")
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("ValiantLabs/Muse-Glimmer-30B-Esper4")
model = AutoModelForMultimodalLM.from_pretrained("ValiantLabs/Muse-Glimmer-30B-Esper4", device_map="auto")
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 ValiantLabs/Muse-Glimmer-30B-Esper4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ValiantLabs/Muse-Glimmer-30B-Esper4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ValiantLabs/Muse-Glimmer-30B-Esper4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/ValiantLabs/Muse-Glimmer-30B-Esper4

SGLang

How to use ValiantLabs/Muse-Glimmer-30B-Esper4 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 "ValiantLabs/Muse-Glimmer-30B-Esper4" \
    --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": "ValiantLabs/Muse-Glimmer-30B-Esper4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "ValiantLabs/Muse-Glimmer-30B-Esper4" \
        --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": "ValiantLabs/Muse-Glimmer-30B-Esper4",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use ValiantLabs/Muse-Glimmer-30B-Esper4 with Docker Model Runner:
```
docker model run hf.co/ValiantLabs/Muse-Glimmer-30B-Esper4
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Support our open-source dataset and model releases!

Esper 4: gemma-4-12B, Qwen3.6-27B

Esper 4 is an agentic coding, architecture, DevOps, and MLOps specialist built on Muse Glimmer 30B!

Your dedicated DevOps expert: Esper 4 maximizes DevOps and architecture helpfulness, powered by high-difficulty DevOps and architecture data generated with DeepSeek-V4-Pro!
Improved coding performance: challenging agentic coding queries allow Esper 4 to tackle harder coding tasks!
AI to build AI: our high-difficulty AI coding and expertise data boosts Esper 4 for AI development, research, deployment, interpretability, operation and experimentation!
Small model sizes allow running on local desktop and mobile, plus super-fast server inference!

Prompting Guide

Esper 4 uses the Muse Glimmer prompt format.

Use Esper 4 with your agentic framework of choice or as a stand-alone chat and code assistant.

Example inference script to get started:

from transformers import AutoProcessor, AutoModelForMultimodalLM

MODEL_ID = "ValiantLabs/Muse-Glimmer-30B-Esper4"
# Load model
processor = AutoProcessor.from_pretrained(MODEL_ID)
model = AutoModelForMultimodalLM.from_pretrained(MODEL_ID, dtype="auto", device_map="auto")

# Prompt
prompt = "Implement CQRS for network appliance config management.\n\nRequirements:\n- Write side: 200 commands/sec, 4 command handlers, SQLite with custom journaling\n- Read side: 1000 queries/sec, 3 read projections in shared memory segments\n- Eventual consistency window: 100ms max\n- Handle atomic swap of projection memory for rebuilds\n- Binary configuration format versioning for schema evolution\n- Framework: libevent with custom protocol parser\n\nConstraints:\n- Manual memory management only, no garbage collection\n- Lock-free data structures where possible\n- Shared memory projections must survive process restarts\n- Command handlers must be thread-safe with 4 worker threads\n- Projection rebuild must not block queries\n- Binary format must support forward/backward compatibility\n- Error handling for corrupted journal recovery\n- Memory-mapped I/O for shared segments\n- Zero-copy where possible for performance\n\nDeliverables:\n1. Command processing pipeline with journaling\n2. Projection engine with shared memory management\n3. Query dispatcher with read-your-writes consistency\n4. Schema evolution system with versioned binary format\n5. Integration with libevent for network I/O\n6. Stress test showing 200 cmd/s + 1000 q/s sustained\n\nAssume x86_64 Linux, pthreads, atomic operations. No high-level frameworks."

messages = [
    {"role": "user", "content": prompt},
]

# Process input
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    add_generation_prompt=True,
    reasoning_strength="high",
).to(model.device)
input_len = inputs["input_ids"].shape[-1]

# Generate output
outputs = model.generate(**inputs)
response = processor.decode(outputs[0][input_len:], skip_special_tokens=False)
print(response)