MetaBase plan
Muse Spark Open
Meta's open-weight reasoning model via the Vercel AI Gateway: a sparse mixture-of-experts (816B total, about 42B active) for coding, agentic tool use, and image understanding with a 1M-token context window. Available at the base tier. Uses 4 base requests per send before length multipliers. Supports image input. Does not support web search.
You always get the exact model you pick — we never silently route you to another.
Specifications
| Provider | Meta |
|---|---|
| Released | 2026-10-15 |
| Intelligence | High |
| Speed | Medium |
| Context window | 1,048,576 tokens |
| Knowledge cutoff | Not officially published |
| Input price | $1.25 / 1M tokens |
| Output price | $4.25 / 1M tokens |
| Request cost | 4 base requests |
| Plan tier | Base |
| Input | Text, Image |
| Output | Text |
| Features | Cached input: $0.15 / 1M tokens, 4 base requests per send before length multipliers, Function calling supported, Image input supported, Served through the Vercel AI Gateway (meta/muse-spark-open), Zero data retention and no prompt training |
| Model ID | muse-spark-open |