Muse Spark Open: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 513ms | 1.36s◆ (best) | 243 tps◆ (best) | 1.0M | $1.25 / $4.25◆ (best) | |
Fireworksfireworks | Healthy | 100%◆ (best) | 87.50% | 90.12% | 90.12% | 512ms◆ (best) | 15.1s | 70 tps | 1.0M | $1.25 / $4.25◆ (best) |
Modalmodal | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 2.54s | 82.6s | 33 tps | 1.0M | $1.25 / $4.25◆ (best) |
| Healthy | 100%◆ (best) | 89.14% | 100%◆ (best) | 100%◆ (best) | 952ms | 7.08s | 112 tps | 1.0M | $1.25 / $4.25◆ (best) | |
Togethertogether | Healthy | 100%◆ (best) | 94.82% | 99.53% | 99.53% | 569ms | 21.4s | 75 tps | 1.0M | $1.25 / $4.25◆ (best) |
- BaseTen100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 513ms
- p99
- 1.36s◆ (best)
- Throughput
- 243 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $1.25 / $4.25◆ (best)
baseten/fp4fp4 - Fireworks100%◆ (best)Healthy24h 87.50%
- 7d
- 90.12%
- 30d
- 90.12%
- p50
- 512ms◆ (best)
- p99
- 15.1s
- Throughput
- 70 tps
- Context
- 1.0M
- $/Mtok
- $1.25 / $4.25◆ (best)
fireworks - Modal100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 2.54s
- p99
- 82.6s
- Throughput
- 33 tps
- Context
- 1.0M
- $/Mtok
- $1.25 / $4.25◆ (best)
modal - Nebius100%◆ (best)Healthy24h 89.14%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 952ms
- p99
- 7.08s
- Throughput
- 112 tps
- Context
- 1.0M
- $/Mtok
- $1.25 / $4.25◆ (best)
nebius/fp4fp4 - Together100%◆ (best)Healthy24h 94.82%
- 7d
- 99.53%
- 30d
- 99.53%
- p50
- 569ms
- p99
- 21.4s
- Throughput
- 75 tps
- Context
- 1.0M
- $/Mtok
- $1.25 / $4.25◆ (best)
together
Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.
Throughput
Median tokens per second.