Meta Muse Spark Open vs Thinking Machines Inkling
Side by side from live data: prices, limits, who serves each model and how reliably, and how they do on our chess arena.
In short
- Inkling is 1.1× cheaper at a typical mix of 3 input tokens per output token.
- Muse Spark Open accepts longer prompts: 1M vs 524K tokens.
- Muse Spark Open's best provider was more reliable over 24 h: 100% vs 98.6%.
- Muse Spark Open is the newer model (released 2026-10-15).
| Muse Spark Open | Inkling | |
|---|---|---|
| Vendor | Meta | Thinking Machines |
| Released | 2026-10-15 | 2026-07-17 |
| Input / 1M | $1.25 | $1.00 |
| Output / 1M | $4.25 | $4.05 |
| Cached input / 1M | $0.15 | $0.17 |
| Blended / 1M (3:1) | $2.00 | $1.76 |
| Context | 1M | 524K |
| Max output | 944K | 472K |
| Accepts | text, image, file, audio, video | text, image, audio |
| Reasoning | Yes | Yes |
| Providers on OpenRouter | 6 | 1 |
| Cheapest provider, output | $4.25 | $4.05 |
| Best provider uptime, 24 h | 100% | 98.6% |
| Chess arena rating | not playing | not playing |
| Vendor API status | — | — |
Cost calculator
Muse Spark Open
$139 per month$4.63 per day
Inkling
$121 per month$4.03 per day
Inkling saves $18.00 a month on this workload.
At list prices per 1M tokens; prompt caching and batch discounts are not included.
Compare Muse Spark Open with