Muna logomuna

Our compiled models.
Your AI client.

We help AI teams run inference with open models, with a focus on boosting GPU utilization so that they serve more and spend less.

Each model is compiled into a self-contained binary and served by an inference runtime we wrote to pack GPUs: several models co-located on a single GPU (megakernel style), with optimizations up and down the stack.

As a result, tokens cost about 40% less here than on OpenRouter, behind the OpenAI and Anthropic APIs you already use.

Quick start

Point the OpenAI SDK at https://inference.muna.ai/v1, or the Anthropic SDK at https://inference.muna.ai, and call any model below.

Get your API key
from openai import OpenAI


# 💥 Create an OpenAI client with the Muna URL
openai = OpenAI(
    base_url="https://inference.muna.ai/v1",
    api_key="<your Muna API key>"
)


# 🔥 Create a chat completion
completion = openai.chat.completions.create(
    model="@google/gemma-4-26b-a4b-it",
    messages=[{ "role": "user", "content": "What is a GPU?" }]
)


# 🚀 Print the output
print(completion.choices[0].message.content)

Models

ModelKindPrice
@google/gemma-4-26b-a4b-it
Stream chat completions from Gemma 4 26B.
chat
$0.065
in / 1M
$0.02
cached / 1M
$0.30
out / 1M
13% cheaper than OpenRouter
@qwen/qwen-3.8-27b
Dense 27B chat model, 128K context, tool calling.
chat
$0.25
in / 1M
$0.02
cached / 1M
$1.75
out / 1M
22% cheaper than OpenRouter
@nomic/nomic-embed-text-v1.5
Text embeddings, 768 dimensions, up to 8K tokens.
embedding
$0.01
/ 1M
@nomic/nomic-embed-text-v1.5
Text embeddings, 768 dimensions, up to 8K tokens.
embedding
$0.01
/ 1M
@qwen/qwen-3-embedding-8b
Multilingual text embeddings, 4,096 dimensions, up to 40K tokens.
embedding
$0.02
/ 1M
@deepseek/deepseek-v4-flash
Fast MoE chat model.Coming soon
chat—
@zai/glm-5.3-flash
Fast chat model from Z.ai.Coming soon
chat—

Need a model that is not here? Ask; bringing one up takes days, not quarters.

How it works

96 MB
compiled Gemma 4 26B binary; the SGLang container is 25.8 GB
6.0 s
cold start to first token
18×
faster cold start than SGLang

It all starts with a Python function. Muna compiles this into a self-contained binary, instead of running a container. The result is a hundred megabytes of code that reads weights off disk and starts serving in seconds.

Because we can spin up a model so quickly, our runtime can load models on demand; co-locate multiple models; and evict when idle. As such, a GPU-hour serves more tokens and the price per token drops.

Cold start to first token · Gemma 4 26B

loading…

Full report with raw data · Methodology