.TXT Serve models?→Release models?↓
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.

← All models
NVIDIA · 8 providers · 366k tool calls a day
#ProviderBroken
tool calls
Tool calls
a day
$ in / out
per M
Tokens/sUptime
1Together#0.66%280k$0.15 / $0.50297100.0%
2io.netnvfp4#1.30%16k$0.15 / $0.506195.0%
3Nebius Token Factorybf16#1.34%4.0k$0.30 / $0.9019296.0%
4Venicefp4#3.44%1.5k$0.20 / $0.5031198.6%
5Crusoebf16#4.58%8.4k$0.10 / $0.4015298.8%
6Basetenfp4#5.18%5.5k$0.20 / $0.5037499.7%
7DeepInfrafp4#7.95%24k$0.15 / $0.5040695.3%
8CoreWeavenvfp4#8.36%27k$0.15 / $0.5035898.5%
Recommended OpenRouter settings for Nemotron 3.5 Super VL
{
  "provider": {
    "order": [
      "together",
      "io-net/nvfp4",
      "nebius/bf16"
    ],
    "ignore": [
      "venice/fp4",
      "crusoe/bf16",
      "baseten/fp4",
      "deepinfra/fp4",
      "coreweave/nvfp4"
    ]
  }
}

Point your agent at llms.txt

Using Nemotron 3.5 Super VL?

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.

Worst provider8.36%

of Nemotron 3.5 Super VL's tool calls fail on CoreWeave (nvfp4).

Best to worst13×

as often on CoreWeave (nvfp4) as on Together, its best provider.

Every day7k

broken tool calls a day across its 8 providers.