Hormouz AIHormouzSign in

Meta: Muse Spark 1.1: AI model

Muse Spark 1.1 is Meta's multimodal reasoning model for agents from Meta with a 1M-token context window. Reachable from inside Iran through Hormouz.

Context window:
1,048,576 tokens
Input:
$1.25/1M tokens
Output:
$4.25/1M tokens
Type:
text, image, video
  • Vision
  • Tools
  • Streaming
  • Long context

Thinking levels: Always thinks · xhigh · high · medium · low · minimal

Muse Spark 1.1 is a Meta reasoning model for agent work. It reads text, images, video and PDFs over a 1M-token context, can lead a multi-agent workflow or run as a sub-agent, and adapts to new tools, MCP servers and custom skills without extra training. It also offers built-in search with citations. Meta reports strong results on coding in large codebases, computer use and turning visuals into code.

It accepts text, image, file (PDF) and video input and returns text, and supports tool calling, structured JSON output and streaming. It always reasons before answering, and you choose how hard it thinks with the reasoning effort setting. Muse Spark 1.1 is commonly used for multi-agent orchestration, coding in large codebases, computer use, visual-to-code and research with citations. Hormouz serves it behind an OpenAI-compatible API: point your base URL at Hormouz, drop in a Hormouz key, and existing OpenAI SDK code runs without changes.

Pricing is usage-based: $1.25 per million input tokens and $4.25 per million output tokens. There is no fixed monthly fee, and you pay only for the tokens you use. Hormouz provides stable access to the model from inside Iran, with no foreign credit card or circumvention tooling required.

Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, and PDF documents and returns text, with a 1M-token context window....

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta/muse-spark-1.1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorMeta
TypeText, Image, Video
Context window1,048,576 tokens
Pricing$1.25/M in · $4.25/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

33.7
Intelligence Index
Better than 63% of models we compared
71.3
Coding Index
Better than 83% of models we compared
25.8
Agentic Index
Better than 59% of models we compared

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
89.8%
HLE
Humanity's Last Exam
46.2%
AA-LCR
Long-context reasoning
77.7%
CritPt
Research-level physics reasoning
15.1%
AA Omniscience
Broad factual knowledge accuracy
52%
Non-Hallucination
Rate of avoiding fabricated answers
50%

Coding

SciCode
Python for scientific computing
58.8%

Agentic

GDPval
Economically valuable tasks
35.7%

Source: Artificial Analysis

Agentic Game Dev
Top 46%
1,183Elo
Win rate: 48.3%
Android Native
Top 49%
1,188Elo
Win rate: 47.2%
Full-Stack
Top 41%
1,198Elo
Win rate: 45.4%
Godot Game Dev
Top 81%
1,104Elo
Win rate: 39.8%
HTML Slides
Top 35%
1,194Elo
Win rate: 51.8%
Mobile Apps
Top 46%
1,174Elo
Win rate: 46.5%
Python Slides
Top 70%
1,156Elo
Win rate: 44.1%
Web Apps
Top 48%
1,199Elo
Win rate: 49.3%
3D
Top 18%
1,270Elo
Win rate: 49.3%Avg time: 197s
ASCII Art
Top 7%
1,289Elo
Win rate: 56.7%Avg time: 168s
Code
Top 14%
1,283Elo
Win rate: 51.4%Avg time: 220s
Data Viz
Top 14%
1,280Elo
Win rate: 50.9%Avg time: 186s
Game Dev
Top 14%
1,285Elo
Win rate: 48.1%Avg time: 227s
SVG
Top 16%
1,240Elo
Win rate: 49.8%Avg time: 124s
UI Component
Top 12%
1,296Elo
Win rate: 51.4%Avg time: 179s
Website
Top 17%
1,276Elo
Win rate: 51.6%Avg time: 218s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meta/muse-spark-1.1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="meta/muse-spark-1.1",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "meta/muse-spark-1.1",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the Meta: Muse Spark 1.1 API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches Meta: Muse Spark 1.1 and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an account

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Muse Spark 1.1 cost?
Pricing is usage-based: $1.25 per million input tokens and $4.25 per million output tokens. There is no fixed monthly fee, and you pay only for the tokens you use.
How do I access Muse Spark 1.1 from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card or circumvention tooling required.
Is Muse Spark 1.1 compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Point your base URL at Hormouz and swap in your Hormouz key. Your existing SDKs and code run without a rewrite.
What context window and capabilities does Muse Spark 1.1 have?
It has a 1M-token context window, accepts text, image, file (PDF) and video input, and supports tool calling, structured JSON output and streaming. It always reasons before answering, and you choose how hard it thinks with the reasoning effort setting.
What is Muse Spark 1.1 good for?
Typical uses are multi-agent orchestration, coding in large codebases, computer use, visual-to-code and research with citations.

Updated 2026-10-13