of Kimi K2.6's tool calls fail on Moonshot AI.
Updated October 29, 2026
Tool-call index
Not all models are served equally by all providers. On a typical model, the worst provider breaks tool calls 19× as often as the best. Find the best one for your agents.
| # | Provider | Broken tool calls | Tool calls a day | $ in / out per M | Tokens/s | Uptime |
|---|---|---|---|---|---|---|
| 1 | SiliconFlowfp8# | 0.06% | 11k | $0.77 / $3.40 | 24 | 100.0% |
| 2 | NovitaAI# | 0.13% | 9.4k | $0.80 / $3.40 | 46 | 100.0% |
| 3 | Inceptronint4# | 0.22% | 52k | $0.62 / $3.50 | 71 | 99.9% |
| 4 | DeepInfrafp4# | 0.51% | 23k | $0.75 / $3.50 | 64 | 99.6% |
| 5 | Parasailint4# | 0.51% | 18k | $0.75 / $3.50 | 77 | 99.9% |
| 6 | DigitalOcean# | 0.77% | 6.4k | $0.57 / $2.40 | 27 | 99.9% |
| 7 | Crusoebf16# | 0.86% | 15k | $0.70 / $3.50 | 72 | 99.2% |
| 8 | AtlasCloudint4# | 0.98% | 1.9k | $0.95 / $4.00 | 83 | 99.9% |
| 9 | Baidu Qianfanfp4# | 1.37% | 38k | $0.39 / $1.64 | 65 | 99.9% |
| 10 | Veniceint4# | 1.41% | 2.6k | $0.75 / $3.50 | 65 | 97.6% |
| 11 | StreamLakefp8# | 1.46% | 1.8k | $0.60 / $2.52 | 43 | 97.1% |
| 12 | CoreWeavefp4# | 1.83% | 69k | $0.65 / $3.41 | 113 | 99.5% |
| 13 | Chutesint4# | 2.07% | 4.5k | $0.39 / $1.80 | 51 | 100.0% |
| 14 | Moonshot AI# | 2.92% | 25k | $0.95 / $4.00 | 76 | 99.6% |
Recommended OpenRouter settings for Kimi K2.6
{
"provider": {
"order": [
"siliconflow/fp8",
"novita",
"inceptron/int4"
],
"ignore": []
}
}Point your agent at llms.txt
Get updates when its successor drops
And which provider to use.
For model makers
Broken tool calls carry your model's name
Releasing a model? Test it before launch.
as often on Moonshot AI as on SiliconFlow (fp8), its best provider.
broken tool calls a day across its 14 providers.