Meta: Muse Image

Muse Image is an agentic image generation model from Meta that generates and edits images from text and reference images. Unlike single-pass image models, it reasons before it renders, breaking down …

meta/muse-image-1.0-eval-20260824

Muse Image is an agentic image generation model from Meta that generates and edits images from text and reference images. Unlike single-pass image models, it reasons before it renders, breaking down multi-part prompts and refining its output within the chain of thought, and invokes web search for factual accuracy on knowledge-intensive prompts.

The model supports text-to-image generation, targeted image editing, multi-image composition, reference-image conditioning for style and subject consistency across a series, and precise text rendering within generated images. Iterative editing works by passing the previous output image back with a new instruction.

Specifications & capabilities

Input
text, image
Output
image
Context window · tokens
65536
Capabilities
imageOutput, multimodal, visionInput
Description source ↗

Evaluations

1–13 / 13

LMArena

1,406.688

source id
lmarena
leaderboard name
LMArena Image Edit Multi Image Edit Agentic
dimension
multimodal
metric
Arena rating
rank
5
sample size
5170
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1394.0836158067032
confidence upper
1419.2923110883617
Original result ↗

LMArena

1,398.666

source id
lmarena
leaderboard name
LMArena Image Edit Multi Image Edit
dimension
multimodal
metric
Arena rating
rank
8
sample size
126209
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1394.572110901655
confidence upper
1402.7596861248392
Original result ↗

LMArena

1,387.443

source id
lmarena
leaderboard name
LMArena Image Edit Agentic
dimension
multimodal
metric
Arena rating
rank
11
sample size
6293
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1376.497704365274
confidence upper
1398.3882431694283
Original result ↗

LMArena

1,288.605

source id
lmarena
leaderboard name
LMArena Text To Image Portraits
dimension
multimodal
metric
Arena rating
rank
11
sample size
10794
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1281.11795167586
confidence upper
1296.0925727372976
Original result ↗

LMArena

1,400.767

source id
lmarena
leaderboard name
LMArena Image Edit Overall
dimension
multimodal
metric
Arena rating
rank
12
sample size
204077
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1397.0322382422087
confidence upper
1404.5019868505326
Original result ↗

LMArena

1,274.868

source id
lmarena
leaderboard name
LMArena Text To Image Photorealistic
dimension
multimodal
metric
Arena rating
rank
12
sample size
21528
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1268.7250030145028
confidence upper
1281.0105790819507
Original result ↗

LMArena

1,266.17

source id
lmarena
leaderboard name
LMArena Text To Image Overall
dimension
multimodal
metric
Arena rating
rank
12
sample size
59855
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1261.7047339028086
confidence upper
1270.6349373659193
Original result ↗

LMArena

1,230.861

source id
lmarena
leaderboard name
LMArena Text To Image Agentic
dimension
multimodal
metric
Arena rating
rank
12
sample size
4336
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1218.800427674591
confidence upper
1242.921510835984
Original result ↗

LMArena

1,291.439

source id
lmarena
leaderboard name
LMArena Text To Image Text Rendering
dimension
multimodal
metric
Arena rating
rank
13
sample size
18382
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1285.0615962435477
confidence upper
1297.8157681600487
Original result ↗

LMArena

1,274.659

source id
lmarena
leaderboard name
LMArena Text To Image Cartoon
dimension
multimodal
metric
Arena rating
rank
13
sample size
22345
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1268.5738185203281
confidence upper
1280.743649956472
Original result ↗

LMArena

1,256.085

source id
lmarena
leaderboard name
LMArena Text To Image Commercial Design
dimension
multimodal
metric
Arena rating
rank
14
sample size
18945
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1249.8774467912388
confidence upper
1262.2934624301797
Original result ↗

LMArena

1,256.171

source id
lmarena
leaderboard name
LMArena Text To Image Art
dimension
multimodal
metric
Arena rating
rank
15
sample size
8155
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1248.3914930186652
confidence upper
1263.9510595241732
Original result ↗

LMArena

1,235.905

source id
lmarena
leaderboard name
LMArena Text To Image 3d Modeling
dimension
multimodal
metric
Arena rating
rank
15
sample size
5126
evaluated at
2026-11-12T00:00:00.000Z
source model id
Muse Image
confidence lower
1226.6742532141914
confidence upper
1245.1352445960722
Original result ↗

Safety assessments

1–1 / 1

Foundation Model Transparency Index 2025 ↗

31 / 100

source id
stanfordFMTI
canonical model id
meta/muse-image-1.0-eval-20260824
taxonomy
transparency
hazard category
transparency
hazard name
Organization transparency
policy area
Meta
assessment type
organization-level disclosure
metric
disclosure_score
result label
31 / 100
sample size
100
updated at
2026-11-09T18:08:05.227Z

Source specification observations

1–2 / 2

openRouter ↗

field
context_length
value
65536
normalized value
65536
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
max_output_tokens
value
58982
normalized value
58982
scope
source-model
source id
openRouter
confidence
1

Original sources

1–6 / 6

Meta · Model reference ↗

OpenRouter · Model page ↗

Epoch AI · Catalog ↗

GitHub · Catalog ↗

LMArena · Catalog ↗

OpenRouter · Catalog ↗

Source links

1–4 / 4

Explore more models

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace