Kimi K3
moonshotai/kimi-k3InputOutput
Compare its 18 providersIncident RSSCompare with Kimi K2.7 Code, Kimi K2.6, Kimi K2.5
CompareKimi K3
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100% | 99.44% | 97.95% | 97.95% | 4.94s | 48.0s | 16 tps | 1.0M | $3.45 / $17.25 |
Fireworksfireworks/fast | Healthy | 100% | 97.63% | 98.42% | 98.42% | 816ms | 15.5s | 91 tps | 1.0M | $4.50 / $22.50 |
| Healthy | 100% | 99.72% | 99.52% | 99.56% | 1.04s | 13.7s | 77 tps | 1.0M | $3.00 / $15.00 | |
| Healthy | 100% | 94.57% | 89.67% | 89.19% | 3.95s | 88.1s | 7.0 tps | 1.0M | $2.80 / $14.00 | |
Togethertogether | Healthy | 99.91% | 98.89% | 95.45% | 96.24% | 1.32s | 9.40s | 53 tps | 1.0M | $3.00 / $15.00 |
| Healthy | 99.88% | 98.59% | 99.05% | 99.11% | 3.97s | 23.9s | 25 tps | 1.0M | $3.00 / $15.00 | |
DigitalOceandigitalocean | Healthy | 99.80% | 99.51% | 98.62% | 98.78% | 2.02s | 50.5s | 28 tps | 1.0M | $3.00 / $15.00 |
| Healthy | 99.56% | 98.64% | 98.62% | 98.58% | 3.08s | 55.6s | 16 tps | 975K | $2.60 / $13.00 | |
Fireworksfireworks | Healthy | 99.17% | 98.03% | 99.08% | 99.04% | 1.80s | 21.7s | 36 tps | 1.0M | $3.00 / $15.00 |
| Healthy | 98.97% | 96.19% | 93.05% | 92.50% | 1.07s | 14.6s | 34 tps | 1.0M | $2.85 / $14.25 | |
Fireworksfireworks/us | Degraded | 96.34% | 97.87% | 94.98% | 95.05% | 1.46s | 28.6s | 46 tps | 1.0M | $3.30 / $16.50 |
| Healthy | 95.24% | 97.99% | 98.09% | 97.87% | 1.63s | 181s | 23 tps | 1.0M | $3.00 / $15.00 | |
| No data | — | 99.33% | 99.61% | 99.25% | 2.52s | 37.4s | 48 tps | 1.0M | $3.00 / $15.00 | |
Makoramakora | Degraded | — | 85.08% | 81.48% | 81.48% | 3.16s | 32.8s | 25 tps | 1.0M | $2.55 / $12.75 |
| No data | — | 92.23% | 92.38% | 89.88% | 3.28s | 16.8s | 7.0 tps | 1.0M | $6.00 / $22.50 | |
| No data | — | 93.75% | 93.19% | 93.19% | 2.19s | 42.3s | 22 tps | 1.0M | $3.00 / $15.00 | |
Phalaphala | No data | — | 94.83% | 93.69% | 94.03% | 13.6s | 105s | 6.0 tps | 1.0M | $3.00 / $15.00 |
Waferwafer | No data | — | — | 66.76% | 80.43% | — | — | — | 1.0M | $3.00 / $15.00 |
- Alibaba100%Healthy24h 99.44%
- 7d
- 97.95%
- 30d
- 97.95%
- p50
- 4.94s
- p99
- 48.0s
- Throughput
- 16 tps
- Context
- 1.0M
- $/Mtok
- $3.45 / $17.25
alibaba - Fireworks100%Healthy24h 97.63%
- 7d
- 98.42%
- 30d
- 98.42%
- p50
- 816ms
- p99
- 15.5s
- Throughput
- 91 tps
- Context
- 1.0M
- $/Mtok
- $4.50 / $22.50
fireworks/fast - Modal100%Healthy24h 99.72%
- 7d
- 99.52%
- 30d
- 99.56%
- p50
- 1.04s
- p99
- 13.7s
- Throughput
- 77 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
modal/mxfp4mxfp4 - Morph100%Healthy24h 94.57%
- 7d
- 89.67%
- 30d
- 89.19%
- p50
- 3.95s
- p99
- 88.1s
- Throughput
- 7.0 tps
- Context
- 1.0M
- $/Mtok
- $2.80 / $14.00
morph/fp4fp4 - Together99.91%Healthy24h 98.89%
- 7d
- 95.45%
- 30d
- 96.24%
- p50
- 1.32s
- p99
- 9.40s
- Throughput
- 53 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
together - Moonshot AI99.88%Healthy24h 98.59%
- 7d
- 99.05%
- 30d
- 99.11%
- p50
- 3.97s
- p99
- 23.9s
- Throughput
- 25 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
moonshotai/mxfp4mxfp4 - DigitalOcean99.80%Healthy24h 99.51%
- 7d
- 98.62%
- 30d
- 98.78%
- p50
- 2.02s
- p99
- 50.5s
- Throughput
- 28 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
digitalocean - Sail Research99.56%Healthy24h 98.64%
- 7d
- 98.62%
- 30d
- 98.58%
- p50
- 3.08s
- p99
- 55.6s
- Throughput
- 16 tps
- Context
- 975K
- $/Mtok
- $2.60 / $13.00
sail-research/fp4fp4 - Fireworks99.17%Healthy24h 98.03%
- 7d
- 99.08%
- 30d
- 99.04%
- p50
- 1.80s
- p99
- 21.7s
- Throughput
- 36 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
fireworks - DeepInfra98.97%Healthy24h 96.19%
- 7d
- 93.05%
- 30d
- 92.50%
- p50
- 1.07s
- p99
- 14.6s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $2.85 / $14.25
deepinfra/bf16bf16 - Fireworks96.34%Degraded24h 97.87%
- 7d
- 94.98%
- 30d
- 95.05%
- p50
- 1.46s
- p99
- 28.6s
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
fireworks/us - Chutes95.24%Healthy24h 97.99%
- 7d
- 98.09%
- 30d
- 97.87%
- p50
- 1.63s
- p99
- 181s
- Throughput
- 23 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
chutes/mxfp4mxfp4 - No data24h 99.33%
- 7d
- 99.61%
- 30d
- 99.25%
- p50
- 2.52s
- p99
- 37.4s
- Throughput
- 48 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
baseten/fp8fp8 - Degraded24h 85.08%
- 7d
- 81.48%
- 30d
- 81.48%
- p50
- 3.16s
- p99
- 32.8s
- Throughput
- 25 tps
- Context
- 1.0M
- $/Mtok
- $2.55 / $12.75
makora - No data24h 92.23%
- 7d
- 92.38%
- 30d
- 89.88%
- p50
- 3.28s
- p99
- 16.8s
- Throughput
- 7.0 tps
- Context
- 1.0M
- $/Mtok
- $6.00 / $22.50
morph/fastfp4 - No data24h 93.75%
- 7d
- 93.19%
- 30d
- 93.19%
- p50
- 2.19s
- p99
- 42.3s
- Throughput
- 22 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
parasail/fp4fp4 - No data24h 94.83%
- 7d
- 93.69%
- 30d
- 94.03%
- p50
- 13.6s
- p99
- 105s
- Throughput
- 6.0 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
phala - No data24h —
- 7d
- 66.76%
- 30d
- 80.43%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00
wafer
Measured 51s ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
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.
Daily availability
One row per endpoint · one dot per day · last 90 days.
worst hour ≥ 95%90–95%< 90%no data
Colour by each endpoint's worst hour of each UTC day · 18 endpoints · 10 days with data, 73 endpoint-days below 90%.
Price changes
- Kimi K3DigitalOceandigitalocean$2.85 → $3.00 / $14.25 → $15.00cache read $0.285 → $0.3002026-08-27 22:55 UTC
- Kimi K3Phalaphala$3.00 / $15.00cache read $1.50 → $0.3002026-08-21 11:00 UTC