of MiniMax M3's tool calls fail on AtlasCloud (fp8).
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 | ModelRun [by Modular]fp8# | 0.27% | 48k | $0.30 / $1.30 | 130 | 99.1% |
| 2 | Venicefp8# | 0.48% | 140k | $0.30 / $1.20 | 132 | 88.5% |
| 3 | NovitaAIfp8# | 0.92% | 22k | $0.30 / $1.20 | 102 | 99.9% |
| 4 | Parasailfp8# | 1.00% | 9.3k | $0.30 / $1.20 | 121 | 99.5% |
| 5 | DeepInfrafp8# | 1.43% | 4.5k | $0.28 / $1.10 | 24 | 99.8% |
| 6 | Together# | 1.56% | 44k | $0.30 / $1.20 | 131 | 99.8% |
| 7 | MiniMax# | 2.20% | 889k | $0.30 / $1.20 | 118 | 99.7% |
| 8 | StreamLakefp8# | 2.23% | 25k | $0.30 / $1.20 | 95 | 99.4% |
| 9 | CoreWeavefp4# | 4.57% | 61k | $0.23 / $0.96 | 176 | 98.2% |
| 10 | GMICloudfp8# | 4.87% | 21k | $0.24 / $0.96 | 55 | 97.2% |
| 11 | AtlasCloudfp8# | 36.08% | 16k | $0.30 / $1.20 | 141 | 99.4% |
Also ignored, under 1,000 tool calls a day: MARA 3.28%.
Recommended OpenRouter settings for MiniMax M3
{
"provider": {
"order": [
"modelrun/fp8",
"venice/fp8",
"novita/fp8"
],
"ignore": [
"mara",
"coreweave/fp4",
"gmicloud/fp8",
"atlas-cloud/fp8"
]
}
}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 AtlasCloud (fp8) as on ModelRun [by Modular] (fp8), its best provider.
broken tool calls a day across its 11 providers.