jishie

Nemotron 3 Ultra

wandb/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B · family nemotron · in: text → out: text · price provider-declared (via models.dev). Capabilities: tools reasoning text.

$0.91
Blended $/1M (3:1)
$0.5
Input $/1M
$2.15
Output $/1M
$2.65
1M in + 1M out
$0.1 −80%
Cache read $/1M
#3930 of 7,714
Cheapest rank
Provider
CoreWeave (wandb)
Kind
text (outputs text)
Context
262,144 tokens
Max output
262,144 tokens
Capabilities
tools reasoning text

Same model, other providers

“Nemotron 3 Ultra” is offered by 3 providers — cheapest-effective first. This one (CoreWeave) is the cheapest.

ProviderInputOutputBlendedContext
CoreWeave (this)$0.5$2.15$0.91262k
Vercel AI Gateway$0.6$2.4$1.051M
DigitalOcean$0.9$1.7$1.10131k

Call it

jishie indexes & prices models; it does not proxy inference.

Most providers are OpenAI-compatible — point base_url at CoreWeave and pass this model id:

curl $BASE_URL/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B","messages":[{"role":"user","content":"hello"}]}'

Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):

← all CoreWeave models · model index · JSON