DeepSeek V3 0324: 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% | 99.89%◆ (best) | 98.26% | 1.73s◆ (best) | 5.51s◆ (best) | 31 tps◆ (best) | 128K | $0.290 / $1.14 | |
| Healthy | 100%◆ (best) | 100% | 99.79% | 98.99% | 1.91s | 6.79s | 23 tps | 164K | $0.250 / $1.00 | |
| No data | — | — | — | 99.03%◆ (best) | — | — | — | 164K | $0.500 / $1.50 | |
| No data | — | — | — | 91.03% | — | — | — | 164K | $0.240 / $0.900◆ (best) | |
| No data | — | — | — | — | — | — | — | 164K | $0.270 / $1.12 |
- GMICloud100%◆ (best)Healthy24h 100%
- 7d
- 99.89%◆ (best)
- 30d
- 98.26%
- p50
- 1.73s◆ (best)
- p99
- 5.51s◆ (best)
- Throughput
- 31 tps◆ (best)
- Context
- 128K
- $/Mtok
- $0.290 / $1.14
gmicloud/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 100%
- 7d
- 99.79%
- 30d
- 98.99%
- p50
- 1.91s
- p99
- 6.79s
- Throughput
- 23 tps
- Context
- 164K
- $/Mtok
- $0.250 / $1.00
siliconflow/fp8fp8 - No data24h —
- 7d
- —
- 30d
- 99.03%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 164K
- $/Mtok
- $0.500 / $1.50
crusoe/bf16bf16 - No data24h —
- 7d
- —
- 30d
- 91.03%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 164K
- $/Mtok
- $0.240 / $0.900◆ (best)
deepinfra/fp4fp4 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 164K
- $/Mtok
- $0.270 / $1.12
novita/fp8fp8
Measured 2m 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.