TokenReply

Loading…

Reload
74
Classic UIClassic
ClassicSign in

Meta

Meta models · 15

llama-4-scout

Meta Llama 4 Scout 17B/16E mixture-of-experts, natively multimodal with an extended 1.31M token context window.

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
meta-llama/Llama-4-Scout-17B-16E-Instruct: 670.3
83 Models · 2026-10-25

Creative Writing v3
meta-llama/Llama-4-Scout-17B-16E-Instruct: 781.1
145 Models · 2026-10-25

WritingBench · Literature & arts
llama-4-scout-17b-16e-instruct: 49.19
54 Models · 2026-10-25

WritingBench · Character design
llama-4-scout-17b-16e-instruct: 53.6
54 Models · 2026-10-25

WritingBench · Plot development
llama-4-scout-17b-16e-instruct: 51.64
54 Models · 2026-10-25

WritingBench · Fan fiction
llama-4-scout-17b-16e-instruct: 54.5
54 Models · 2026-10-25

4.1/ 10
Input / 1M tokens
$0.25

muse-spark-1.2

Reasoning model from Meta for complex agentic and long-horizon coding work: text, image, video, audio and PDF input, text output, 1M-token context. Supports configurable reasoning effort, structured output and parallel tool calling.

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
muse-spark-1.2: 1,842.3
145 Models · 2026-10-25

LiveBench · Story constraint following
muse-spark-1.2-xhigh: 73.65
71 Models · 2026-10-25

LiveBench · Theory of mind
muse-spark-1.2-xhigh: 80.77
71 Models · 2026-10-25

8.1/ 10
Input / 1M tokens
$1.25

muse-spark

Muse Spark via the Muse web agent: text in, text out. Tool calling and reasoning controls are not supported.

Roleplay

No matching public benchmark is available for this model.

Not measured
Input / 1M tokens
$1.25

muse-spark-plus-free

Plus-only free model. Same model as the paid muse-spark, billed at zero, 50 requests per day.

Roleplay

No matching public benchmark is available for this model.

Not measured
Input / 1M tokens
$0.00

muse-spark-1.3-contributor

Muse Spark 1.3 Contributor

Roleplay

No matching public benchmark is available for this model.

Not measured
Input / 1M tokens
$0.1

muse-image

Muse text-to-image, about 2K (2048x1152 at 16:9). One image per request (n=1); size takes 1:1, 16:9, 9:16, 4:3, 3:4. Flat per-image price.

Roleplay

No matching public benchmark is available for this model.

Not measured
Per request
$0.04

muse-video

Muse text-to-video and first-frame image-to-video on /v1/videos (async task). Billed per requested second (default 4). Output length is decided upstream, typically about 6 s; 16:9 or 9:16.

Roleplay

No matching public benchmark is available for this model.

Not measured
Per request
$0.03

muse-spark-1.2-contributor

Meta Muse Spark 1.2 contributor tier: same weights as muse-spark-1.2 with discounted token pricing in exchange for allowing Meta to train on prompts and completions. 1M context. Not available in all regions.

Roleplay

No matching public benchmark is available for this model.

Not measured
Input / 1M tokens
$0.1
Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
meta-models/Muse-Glimmer-30B: 1,804.5
145 Models · 2026-10-25

8.3/ 10
Per request
$0.00

llama-4-maverick

Meta Llama 4 Maverick 17B/128E mixture-of-experts, natively multimodal with a 1M token context window for chat, vision and tool use.

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
meta-llama/Llama-4-Maverick-17B-128E-Instruct: 864.8
83 Models · 2026-10-25

Creative Writing v3
meta-llama/Llama-4-Maverick-17B-128E-Instruct: 858
145 Models · 2026-10-25

WritingBench · Literature & arts
llama-4-maverick-17b-128e: 51.25
54 Models · 2026-10-25

WritingBench · Character design
llama-4-maverick-17b-128e: 54.8
54 Models · 2026-10-25

WritingBench · Plot development
llama-4-maverick-17b-128e: 51.45
54 Models · 2026-10-25

WritingBench · Fan fiction
llama-4-maverick-17b-128e: 51.5
54 Models · 2026-10-25

4.4/ 10
Input / 1M tokens
$0.35

llama-3.3-70b-instruct

Meta Llama 3.3 70B multilingual instruction-tuned text model with a 128K token context window.

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

WritingBench · Literature & arts
llama-3.3-70b-instruct: 46.04
54 Models · 2026-10-25

WritingBench · Character design
llama-3.3-70b-instruct: 50
54 Models · 2026-10-25

WritingBench · Plot development
llama-3.3-70b-instruct: 48.36
54 Models · 2026-10-25

WritingBench · Fan fiction
llama-3.3-70b-instruct: 49
54 Models · 2026-10-25

4.2/ 10
Input / 1M tokens
$0.25

muse-glimmer-30b

Dense open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for autonomous agents on consumer hardware: text and image input, text output, 128K context, configurable reasoning effort.

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
meta-models/Muse-Glimmer-30B: 1,804.5
145 Models · 2026-10-25

8.3/ 10
Input / 1M tokens
$0.35

muse-spark-1.3

Muse Spark 1.3

Roleplay

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
muse-spark-1.3: 1,907
145 Models · 2026-10-25

LiveBench · Story constraint following
muse-spark-1.3-xhigh: 77.48
71 Models · 2026-10-25

LiveBench · Theory of mind
muse-spark-1.3-xhigh: 84.62
71 Models · 2026-10-25

8.7/ 10
Input / 1M tokens
$1.25