Model specification · Meta
Muse Spark 1.1
meta/muse-spark-1.1
Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with a 1M-token context...
SpecificationsReported values
- Context window
- 1.0M
- Maximum output
- —
- Architecture
- text+image+file+audio+video->text
- Tokenizer
- Other
- Knowledge cutoff
- —
- Moderated
- Yes
- Input modalities
- text image video file audio
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard |
|---|---|
| Input | $1.25 / M |
| Output | $4.25 / M |
Benchmark indexesReported indexes
- Intelligence
- 50.6
- Coding
- 71.3
- Agentic
- 37.5
- Design Arena Elo
- 1266
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.017 |
| Long analysis | 100K / 4K × 1 | 50% | $0.087 |
| Workday | 20K / 2K × 1,000 | 30% | $26.90 |