Kimi K3
moonshotai/kimi-k3
Compare with Kimi K2.7 Code, Kimi K2.6, Kimi K2.5
CompareKimi K3
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 99.28% | 97.21% | 97.21% | 1.67s | 12.0s | 18 tps | 1.0M | $3.00 / $15.00 | |
DigitalOceandigitalocean | Healthy | 99.91% | 99.46% | 99.31% | 99.31% | 3.34s | 30.0s | 26 tps | 1.0M | $2.85 / $14.25 |
| Healthy | 99.89% | 99.55% | 99.65% | 99.65% | 1.61s | 15.1s | 69 tps | 1.0M | $3.00 / $15.00 | |
| Healthy | 99.74% | 95.91% | 88.21% | 88.21% | 3.56s | 40.6s | 21 tps | 1.0M | $2.80 / $14.00 | |
Fireworksfireworks | Healthy | 98.95% | 98.44% | 98.93% | 98.93% | 3.14s | 44.4s | 25 tps | 1.0M | $3.00 / $15.00 |
Togethertogether | Healthy | 98.21% | 98.90% | 98.72% | 98.72% | 2.67s | 26.0s | 34 tps | 1.0M | $3.00 / $15.00 |
Fireworksfireworks/us | Healthy | 97.91% | 92.89% | 95.44% | 95.44% | 1.23s | 21.5s | 56 tps | 1.0M | $3.30 / $16.50 |
| Healthy | 97.80% | 97.40% | 98.48% | 98.48% | 4.12s | 108s | 28 tps | 975K | $2.60 / $13.00 | |
| Degraded | 94.65% | 98.43% | 99.15% | 99.15% | 5.20s | 48.4s | 22 tps | 1.0M | $3.00 / $15.00 | |
Phalaphala | Down | 51.11% | 95.67% | 94.92% | 94.92% | 4.04s | 134s | 28 tps | 1.0M | $3.00 / $15.00 |
| No data | — | 98.44% | 98.25% | 98.25% | 469ms | 32.2s | 71 tps | 1.0M | $3.00 / $15.00 | |
| Degraded | — | 91.09% | 88.82% | 88.82% | 1.09s | 27.5s | 17 tps | 1.0M | $2.85 / $14.25 | |
| No data | — | 84.47% | 82.82% | 82.82% | 3.22s | 101s | 23 tps | 1.0M | $6.00 / $22.50 | |
Waferwafer | No data | — | 97.05% | 96.96% | 96.96% | 37.3s | 37.3s | 10 tps | 1.0M | $3.00 / $15.00 |
- Chutes100%Healthy24h 99.28%
- 7d
- 97.21%
- 30d
- 97.21%
- p50
- 1.67s
- p99
- 12.0s
- Throughput
- 18 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
chutes/mxfp4mxfp4 - DigitalOcean99.91%Healthy24h 99.46%
- 7d
- 99.31%
- 30d
- 99.31%
- p50
- 3.34s
- p99
- 30.0s
- Throughput
- 26 tps
- Context
- 1.0M
- $/Mtok
- $2.85 / $14.25
digitalocean - Modal99.89%Healthy24h 99.55%
- 7d
- 99.65%
- 30d
- 99.65%
- p50
- 1.61s
- p99
- 15.1s
- Throughput
- 69 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
modal/mxfp4mxfp4 - Morph99.74%Healthy24h 95.91%
- 7d
- 88.21%
- 30d
- 88.21%
- p50
- 3.56s
- p99
- 40.6s
- Throughput
- 21 tps
- Context
- 1.0M
- $/Mtok
- $2.80 / $14.00
morph/fp4fp4 - Fireworks98.95%Healthy24h 98.44%
- 7d
- 98.93%
- 30d
- 98.93%
- p50
- 3.14s
- p99
- 44.4s
- Throughput
- 25 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
fireworks - Together98.21%Healthy24h 98.90%
- 7d
- 98.72%
- 30d
- 98.72%
- p50
- 2.67s
- p99
- 26.0s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
together - Fireworks97.91%Healthy24h 92.89%
- 7d
- 95.44%
- 30d
- 95.44%
- p50
- 1.23s
- p99
- 21.5s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
fireworks/us - Sail Research97.80%Healthy24h 97.40%
- 7d
- 98.48%
- 30d
- 98.48%
- p50
- 4.12s
- p99
- 108s
- Throughput
- 28 tps
- Context
- 975K
- $/Mtok
- $2.60 / $13.00
sail-research/fp4fp4 - Moonshot AI94.65%Degraded24h 98.43%
- 7d
- 99.15%
- 30d
- 99.15%
- p50
- 5.20s
- p99
- 48.4s
- Throughput
- 22 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
moonshotai/mxfp4mxfp4 - Phala51.11%Down24h 95.67%
- 7d
- 94.92%
- 30d
- 94.92%
- p50
- 4.04s
- p99
- 134s
- Throughput
- 28 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
phala - No data24h 98.44%
- 7d
- 98.25%
- 30d
- 98.25%
- p50
- 469ms
- p99
- 32.2s
- Throughput
- 71 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
baseten/fp8fp8 - Degraded24h 91.09%
- 7d
- 88.82%
- 30d
- 88.82%
- p50
- 1.09s
- p99
- 27.5s
- Throughput
- 17 tps
- Context
- 1.0M
- $/Mtok
- $2.85 / $14.25
deepinfra/bf16bf16 - No data24h 84.47%
- 7d
- 82.82%
- 30d
- 82.82%
- p50
- 3.22s
- p99
- 101s
- Throughput
- 23 tps
- Context
- 1.0M
- $/Mtok
- $6.00 / $22.50
morph/fastfp4 - No data24h 97.05%
- 7d
- 96.96%
- 30d
- 96.96%
- p50
- 37.3s
- p99
- 37.3s
- Throughput
- 10 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
wafer
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
3 endpoints reporting no data are omitted from the charts.
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.