Kimi K3
moonshotai/kimi-k3
Compare with Kimi K2.7 Code, Kimi K2.6, Kimi K2.5
CompareKimi K3
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 98.40% | 98.19% | 98.19% | 685ms | 26.0s | 77 tps | 1.0M | $3.00 / $15.00 | |
| Healthy | 100% | 99.27% | 97.14% | 97.14% | 1.51s | 10.3s | 24 tps | 1.0M | $3.00 / $15.00 | |
Fireworksfireworks | Healthy | 100% | 98.39% | 98.92% | 98.92% | 2.93s | 22.1s | 36 tps | 1.0M | $3.00 / $15.00 |
DigitalOceandigitalocean | Healthy | 99.91% | 99.45% | 99.30% | 99.30% | 3.07s | 29.1s | 24 tps | 1.0M | $2.85 / $14.25 |
| Degraded | 99.81% | 98.68% | 99.30% | 99.30% | 5.93s | 54.6s | 22 tps | 1.0M | $3.00 / $15.00 | |
| Healthy | 99.71% | 99.55% | 99.65% | 99.65% | 1.98s | 19.5s | 74 tps | 1.0M | $3.00 / $15.00 | |
| Healthy | 99.11% | 94.89% | 87.79% | 87.79% | 3.26s | 42.0s | 21 tps | 1.0M | $2.80 / $14.00 | |
Togethertogether | Healthy | 98.94% | 98.97% | 98.77% | 98.77% | 2.43s | 22.3s | 49 tps | 1.0M | $3.00 / $15.00 |
| Healthy | 97.29% | 97.49% | 98.50% | 98.50% | 3.74s | 82.9s | 27 tps | 975K | $2.60 / $13.00 | |
Fireworksfireworks/us | Healthy | 95.97% | 92.83% | 95.41% | 95.41% | 830ms | 9.48s | 64 tps | 1.0M | $3.30 / $16.50 |
Phalaphala | Down | 80.92% | 96.27% | 95.10% | 95.10% | 2.93s | 18.5s | 40 tps | 1.0M | $3.00 / $15.00 |
| Down | 73.86% | 92.23% | 89.41% | 89.41% | 893ms | 22.1s | 18 tps | 1.0M | $2.85 / $14.25 | |
| No data | — | 84.22% | 82.82% | 82.82% | 4.69s | 12.3s | 22 tps | 1.0M | $6.00 / $22.50 | |
Waferwafer | No data | — | 97.37% | 96.96% | 96.96% | — | — | — | 1.0M | $3.00 / $15.00 |
- BaseTen100%Healthy24h 98.40%
- 7d
- 98.19%
- 30d
- 98.19%
- p50
- 685ms
- p99
- 26.0s
- Throughput
- 77 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
baseten/fp8fp8 - Chutes100%Healthy24h 99.27%
- 7d
- 97.14%
- 30d
- 97.14%
- p50
- 1.51s
- p99
- 10.3s
- Throughput
- 24 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
chutes/mxfp4mxfp4 - Fireworks100%Healthy24h 98.39%
- 7d
- 98.92%
- 30d
- 98.92%
- p50
- 2.93s
- p99
- 22.1s
- Throughput
- 36 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
fireworks - DigitalOcean99.91%Healthy24h 99.45%
- 7d
- 99.30%
- 30d
- 99.30%
- p50
- 3.07s
- p99
- 29.1s
- Throughput
- 24 tps
- Context
- 1.0M
- $/Mtok
- $2.85 / $14.25
digitalocean - Moonshot AI99.81%Degraded24h 98.68%
- 7d
- 99.30%
- 30d
- 99.30%
- p50
- 5.93s
- p99
- 54.6s
- Throughput
- 22 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
moonshotai/mxfp4mxfp4 - Modal99.71%Healthy24h 99.55%
- 7d
- 99.65%
- 30d
- 99.65%
- p50
- 1.98s
- p99
- 19.5s
- Throughput
- 74 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
modal/mxfp4mxfp4 - Morph99.11%Healthy24h 94.89%
- 7d
- 87.79%
- 30d
- 87.79%
- p50
- 3.26s
- p99
- 42.0s
- Throughput
- 21 tps
- Context
- 1.0M
- $/Mtok
- $2.80 / $14.00
morph/fp4fp4 - Together98.94%Healthy24h 98.97%
- 7d
- 98.77%
- 30d
- 98.77%
- p50
- 2.43s
- p99
- 22.3s
- Throughput
- 49 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
together - Sail Research97.29%Healthy24h 97.49%
- 7d
- 98.50%
- 30d
- 98.50%
- p50
- 3.74s
- p99
- 82.9s
- Throughput
- 27 tps
- Context
- 975K
- $/Mtok
- $2.60 / $13.00
sail-research/fp4fp4 - Fireworks95.97%Healthy24h 92.83%
- 7d
- 95.41%
- 30d
- 95.41%
- p50
- 830ms
- p99
- 9.48s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
fireworks/us - Phala80.92%Down24h 96.27%
- 7d
- 95.10%
- 30d
- 95.10%
- p50
- 2.93s
- p99
- 18.5s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
phala - DeepInfra73.86%Down24h 92.23%
- 7d
- 89.41%
- 30d
- 89.41%
- p50
- 893ms
- p99
- 22.1s
- Throughput
- 18 tps
- Context
- 1.0M
- $/Mtok
- $2.85 / $14.25
deepinfra/bf16bf16 - No data24h 84.22%
- 7d
- 82.82%
- 30d
- 82.82%
- p50
- 4.69s
- p99
- 12.3s
- Throughput
- 22 tps
- Context
- 1.0M
- $/Mtok
- $6.00 / $22.50
morph/fastfp4 - No data24h 97.37%
- 7d
- 96.96%
- 30d
- 96.96%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
wafer
Measured 5m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
2 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.