Alibaba
99.90%Context1M
Max output1M
p50 latency783 ms
Throughput60.0 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Model directory
Baseer model profileLanguage model
moonshotai/kimi-k3
Kimi K3
Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window.
Input / 1M tokens$2.5
Output / 1M tokens$12.75
Maximum context1M
Inference providers14
Price comparison
USD per one million tokens, using current provider data.
Provider latency
Last-hour p50 in milliseconds; lower is better.
Provider uptime
Reported availability over the last 24 hours; not a contractual SLA.
Inference providers
Compare current pricing, performance, and privacy across every available provider.
Baseten
96.50%Context1M
Max output131.1K
p50 latency1,391 ms
Throughput90.0 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Blackbox
99.95%Context1M
Max output131.1K
p50 latency1,082 ms
Throughput120.5 t/s
Input price$3
Output price$15
Caching
Deepinfra
97.82%Context1M
Max output131.1K
p50 latency407 ms
Throughput96.5 t/s
Input price$2.85
Output price$14.25
Zero retentionNo trainingCaching
Digitalocean
99.96%Context1M
Max output131.1K
p50 latency566 ms
Throughput81.0 t/s
Input price$2.85
Output price$14.25
Zero retentionNo trainingCaching
Fireworks
100.00%Context1M
Max output131.1K
p50 latency2,267 ms
Throughput70.0 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Modal
99.89%Context1M
Max output1M
p50 latency616 ms
Throughput174.5 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Moonshotai
100.00%Context1M
Max output131.1K
p50 latency1,712 ms
Throughput33.0 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Morph
100.00%Context1M
Max output1M
p50 latency2,118 ms
Throughput16.5 t/s
Input price$2.5
Output price$14
Zero retentionNo trainingCaching
Nebius
98.40%Context1M
Max output131.1K
p50 latency1,329 ms
Throughput23.0 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Parasail
99.98%Context1M
Max output1M
p50 latency493 ms
Throughput84.5 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Runware
99.73%Context1M
Max output1M
p50 latency503 ms
Throughput73.0 t/s
Input price$3
Output price$15
Caching
Togetherai
95.85%Context1M
Max output131.1K
p50 latency620 ms
Throughput81.5 t/s
Input price$3
Output price$15
Zero retentionNo trainingCaching
Wafer
99.90%Context1M
Max output131.1K
p50 latency504 ms
Throughput65.0 t/s
Input price$3
Output price$12.75
Zero retentionNo training
| Provider | Maximum context | Max output | p50 latency | p50 throughput | 24h uptime | Input / 1M tokens | Output / 1M tokens | Privacy |
|---|---|---|---|---|---|---|---|---|
Alibaba | 1M | 1M | 783 ms | 60.0 t/s | 99.897% | $3 | $15 | |
Baseten | 1M | 131.1K | 1,391 ms | 90.0 t/s | 96.498% | $3 | $15 | |
Blackbox | 1M | 131.1K | 1,082 ms | 120.5 t/s | 99.951% | $3 | $15 | |
Deepinfra | 1M | 131.1K | 407 ms | 96.5 t/s | 97.816% | $2.85 | $14.25 | |
Digitalocean | 1M | 131.1K | 566 ms | 81.0 t/s | 99.962% | $2.85 | $14.25 | |
Fireworks | 1M | 131.1K | 2,267 ms | 70.0 t/s | 99.998% | $3 | $15 | |
Modal | 1M | 1M | 616 ms | 174.5 t/s | 99.890% | $3 | $15 | |
Moonshotai | 1M | 131.1K | 1,712 ms | 33.0 t/s | 100.000% | $3 | $15 | |
Morph | 1M | 1M | 2,118 ms | 16.5 t/s | 100.000% | $2.5 | $14 | |
Nebius | 1M | 131.1K | 1,329 ms | 23.0 t/s | 98.399% | $3 | $15 | |
Parasail | 1M | 1M | 493 ms | 84.5 t/s | 99.977% | $3 | $15 | |
Runware | 1M | 1M | 503 ms | 73.0 t/s | 99.727% | $3 | $15 | |
Togetherai | 1M | 131.1K | 620 ms | 81.5 t/s | 95.853% | $3 | $15 | |
Wafer | 1M | 131.1K | 504 ms | 65.0 t/s | 99.898% | $3 | $12.75 |
What can this model do?
Available modalities and capabilities across inference providers.
TextImageFileAdvanced reasoningTool useAutomatic cachingFile inputVisionVideo inputTools
Reasoning options
LowHighMax
Technical profile
- Model type
- Language model
- Release date
- Jul 16, 2026
- Knowledge cutoff
- —
- Regions
- US
- API specifications
- V2, V3, V4
- Temperature control
- Not supported
Model identifier
moonshotai/kimi-k3Model profile updated: Sep 12, 2026, 1:46 PM
Related models
More options from the same creator or model category.