of Nemotron 3.5 Super VL's tool calls fail on CoreWeave (nvfp4).
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 | Together# | 0.66% | 280k | $0.15 / $0.50 | 297 | 100.0% |
| 2 | io.netnvfp4# | 1.30% | 16k | $0.15 / $0.50 | 61 | 95.0% |
| 3 | Nebius Token Factorybf16# | 1.34% | 4.0k | $0.30 / $0.90 | 192 | 96.0% |
| 4 | Venicefp4# | 3.44% | 1.5k | $0.20 / $0.50 | 311 | 98.6% |
| 5 | Crusoebf16# | 4.58% | 8.4k | $0.10 / $0.40 | 152 | 98.8% |
| 6 | Basetenfp4# | 5.18% | 5.5k | $0.20 / $0.50 | 374 | 99.7% |
| 7 | DeepInfrafp4# | 7.95% | 24k | $0.15 / $0.50 | 406 | 95.3% |
| 8 | CoreWeavenvfp4# | 8.36% | 27k | $0.15 / $0.50 | 358 | 98.5% |
{
"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
And which provider to use.
Releasing a model? Test it before launch.
of Nemotron 3.5 Super VL's tool calls fail on CoreWeave (nvfp4).
as often on CoreWeave (nvfp4) as on Together, its best provider.
broken tool calls a day across its 8 providers.