Muse Glimmer 30B: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Phalaphala | Healthy | 99.67%◆ (best) | 99.91% | 98.33% | 99.23% | 3.05s | 10.2s | 28 tps◆ (best) | 131K | $0.300 / $1.10◆ (best) |
Togethertogether | Healthy | 98.28% | 98.15% | 99.58%◆ (best) | 99.40% | 2.25s◆ (best) | 7.49s◆ (best) | 25 tps | 131K | $0.350 / $1.50 |
| Degraded | 95.07% | 93.59% | 99.02% | 99.74%◆ (best) | 16.6s | 36.9s | 6.0 tps | 131K | $0.300 / $1.20◆ (best) | |
Fireworksfireworks | No data | — | — | — | 98.44% | — | — | — | 131K | $0.350 / $1.50 |
| No data | — | — | — | — | — | — | — | 131K | $0.300 / $1.10◆ (best) |
- Phala99.67%◆ (best)Healthy24h 99.91%
- 7d
- 98.33%
- 30d
- 99.23%
- p50
- 3.05s
- p99
- 10.2s
- Throughput
- 28 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.300 / $1.10◆ (best)
phala - Together98.28%Healthy24h 98.15%
- 7d
- 99.58%◆ (best)
- 30d
- 99.40%
- p50
- 2.25s◆ (best)
- p99
- 7.49s◆ (best)
- Throughput
- 25 tps
- Context
- 131K
- $/Mtok
- $0.350 / $1.50
together - DeepInfra95.07%Degraded24h 93.59%
- 7d
- 99.02%
- 30d
- 99.74%◆ (best)
- p50
- 16.6s
- p99
- 36.9s
- Throughput
- 6.0 tps
- Context
- 131K
- $/Mtok
- $0.300 / $1.20◆ (best)
deepinfra/bf16bf16 - No data24h —
- 7d
- —
- 30d
- 98.44%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.350 / $1.50
fireworks - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.300 / $1.10◆ (best)
parasail/bf16bf16
Measured 3m 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.