Muse Spark Open
meta-llama/muse-spark-open
Meta· 2026-10-15· 1.0M ctx
API onlytool callreasoningstructured quant: unknown
spread (output, p90 ÷ p10)
1.0×
$4.25 – $5.31 · 5 providers · max÷min 1.3×
5 providers priced · 5 offerings
Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.
Blended $/M assumptions
blended = 90.9% input × [(1 − 70% cache) × input + 70% cache × cache-read] + 9.1% × output + cache writes (30% of uncached input) where a write price is published
A provider that publishes no cache price is blended with cache share 0 (all input at the input price) and labelled “no cache price published” — never penalised against identical in/out prices.
| Provider | Input | Output | Cache rd | Blended |
|---|---|---|---|---|
Vercel AI Gateway router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
Eden AI router / aggregator | $1.25 | $4.25 | $1.25 | $1.52 |
Fireworks AI inference host | $1.25 | $4.25 | $0.15 | $0.82 |
Kilo Gateway router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
Merge Gateway router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
NanoGPT router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
Nebius Token Factory inference host | $1.25 | $4.25 | $1.25 | $1.52 |
Ollama Cloud inference host | $1.25 | $4.25 | $0.15 | $0.82 |
OpenRouter router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
Prime Inference unclassified | $1.25 | $4.25 | — | $1.52no cache price published |
Requesty router / aggregator | $1.25 | $4.25 | $0.15 | $0.82 |
Together AI inference host | $1.25 | $4.25 | $0.15 | $0.82 |
Venice AIoutlier inference host | $1.56 | $5.31 | $0.19 | $1.03 |
Data: models.dev (MIT), snapshot 2026-10-17. Prices are vendor-published; verify before purchase. Methodology.