modelgrep
M

Meta: Muse Spark 1.2

meta/muse-spark-1.2

8th smartest of 158ReasoningToolsJSONVisionAudio
Use via OpenRouter ↗
Intelligence
54.1
8th of 158
Design Elo
Speed
149
36th fastest
Latency
3.7s
first token
Input price
$1.25
222nd cheapest
Context
1.0M

How it compares

Smarter than95%
of all ranked models
Faster than89%
of all ranked models
Cheaper than27%
of all ranked models

Overview

Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, audio, and PDF documents, returns text, and offers a 1M-token context...

Benchmarks

independent · Artificial Analysis & Design Arena
Artificial Analysis98th percentile
Intelligence Index
54.1
Coding Index
72.2
GPQA Diamond
90%
Humanity's Last Exam
44%
SciCode
56%
Tau²-Bench (agentic)
27%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Meta$1.25$4.25100%

Specifications

Context window1.0M
Max output
Knowledge cutoff
Input modalitiestext, image, video, file, audio
Output modalitiestext
Prompt caching
Cache read price$0.150/M
ModeratedYes

Muse Spark 1.2 FAQ

How much does Muse Spark 1.2 cost?

Muse Spark 1.2 costs $1.25 per million input tokens and $4.25 per million output tokens via OpenRouter, making it 222nd cheapest of 306 paid models.

How smart is Muse Spark 1.2?

Muse Spark 1.2 scores 54.1 on the Artificial Analysis Intelligence Index, ranking 8th of 158 benchmarked models, with a GPQA Diamond score of 90%.

How fast is Muse Spark 1.2?

Muse Spark 1.2 generates around 149 tokens per second with 3.7s time-to-first-token (p50), the 36th fastest tracked model.

What is Muse Spark 1.2's context window?

Muse Spark 1.2 supports a 1.0M-token context window. It accepts text, image, video, file, audio input.

Compare head-to-head