DeepSeek V3 0324
deepseek/deepseek-chat-v3-0324
Compare with DeepSeek V3, DeepSeek V3.1, DeepSeek V3.1 Terminus
CompareDeepSeek V3 0324
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 99.95% | 99.94% | 99.94% | 416ms | 7.51s | 33 tps | 164K | $0.500 / $1.50 | |
| Healthy | 99.52% | 99.04% | 98.73% | 98.73% | 3.06s | 7.47s | 22 tps | 164K | $0.290 / $1.14 | |
| Healthy | 99.02% | 97.24% | 97.38% | 97.38% | 1.45s | 9.95s | 30 tps | 164K | $0.270 / $1.12 | |
| Degraded | 93.37% | 97.12% | 96.63% | 96.63% | 1.58s | 12.8s | 19 tps | 164K | $0.250 / $1.00 | |
| Down | 24.24% | 84.55% | 76.88% | 76.88% | 7.55s | 138s | 17 tps | 164K | $0.240 / $0.900 |
- Crusoe100%Healthy24h 99.95%
- 7d
- 99.94%
- 30d
- 99.94%
- p50
- 416ms
- p99
- 7.51s
- Throughput
- 33 tps
- Context
- 164K
- $/Mtok
- $0.500 / $1.50
crusoe/bf16bf16 - GMICloud99.52%Healthy24h 99.04%
- 7d
- 98.73%
- 30d
- 98.73%
- p50
- 3.06s
- p99
- 7.47s
- Throughput
- 22 tps
- Context
- 164K
- $/Mtok
- $0.290 / $1.14
gmicloud/fp8fp8 - Novita99.02%Healthy24h 97.24%
- 7d
- 97.38%
- 30d
- 97.38%
- p50
- 1.45s
- p99
- 9.95s
- Throughput
- 30 tps
- Context
- 164K
- $/Mtok
- $0.270 / $1.12
novita/fp8fp8 - SiliconFlow93.37%Degraded24h 97.12%
- 7d
- 96.63%
- 30d
- 96.63%
- p50
- 1.58s
- p99
- 12.8s
- Throughput
- 19 tps
- Context
- 164K
- $/Mtok
- $0.250 / $1.00
siliconflow/fp8fp8 - DeepInfra24.24%Down24h 84.55%
- 7d
- 76.88%
- 30d
- 76.88%
- p50
- 7.55s
- p99
- 138s
- Throughput
- 17 tps
- Context
- 164K
- $/Mtok
- $0.240 / $0.900
deepinfra/fp4fp4
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
Price changes
- DeepSeek V3 0324GMICloudgmicloud/fp8free / free → $0.290 / $1.142026-08-19 14:25 UTC
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.
Throughput
Median tokens per second.